Electrodeposition of Nb, Ta, Zr and Cu from Ionic Liquid...

Electrodeposition of Nb, Ta, Zr and

Cu from Ionic Liquid for

Nanocomposites Preparation

LAURA MAIS

Dottorato di Ricerca in Ingegneria Industriale

Università degli Studi di Cagliari

XXVII Ciclo

Electrodeposition of Nb, Ta, Zr and

Cu from Ionic Liquid for

Nanocomposites Preparation

LAURA MAIS

Supervisor:

Professor Michele MASCIA

University of Cagliari, Italy

Co-supervisors:

Professor Carlos PONCE DE LEON ALBARRAN

University of Southampton, UK

Dottorato di Ricerca in Ingegneria Industriale

Università degli Studi di Cagliari

XXVII Ciclo

Questa tesi può essere utilizzata, nei limiti stabiliti dalla normativa vigente

sul Diritto d’Autore (Legge 22 aprile 1941 n.633 e succ. modificazioni e articoli

da 2575 a 2583 del Codice civile) ed esclusivamente per scopi didattici e di

ricerca; è vietato qualsiasi utilizzo per fini commerciali. In ogni caso tutti gli

utilizzi devono riportare la corretta citazione delle fonti. La traduzione,

l’adattamento totale e parziale, sono riservati per tutti i Paesi. I documenti

depositati sono sottoposti alla legislazione italiana in vigore nel rispetto del

Diritto di Autore, da qualunque luogo essi siano fruiti.

Laura Mais gratefully acknowledges Sardinia Regional Government

for the financial support of her PhD scholarship (P.O.R. Sardegna

F.S.E. Operational Programme of the Autonomous Region of

Sardinia, European Social Fund 2007-2013 - Axis IV Human

Resources, Objective l.3, Line of Activity l.3.1.

This activity is supported by the European Community Framework

Programme 7, Multiscale Modelling and Materials by Design of

interface-controlled Radiation Damage in Crystalline Materials

(RADINTERFACES), under grant agreement n. 263273.

Acknowledgements

First of all, I would like to thank my supervisor, Prof. Michele

Mascia, for the possibility of undertaking this PhD, and for his

support and advice during these three years at the Department of

Mechanical, Chemical Engineering and Materials (DIMCM),

University of Cagliari, Italy.

I wish to thank Prof. Annalisa Vacca and Prof. Simonetta Palmas for

their support and advices for my research.

I would also like to thank my English supervisor, Prof. Carlos Ponce

de Leon Albarran, for his help during my stay at the University of

Southampton – Engineering Science (UK).

Thanks to my colleagues at the Department of Mechanical, Chemical

Engineering and Materials, for the wonderful time spent together in

the office and for being good friends.

Special thanks to Mario for the love and joy he give me during this

experience we spent together in our life.

Last, I am grateful to my family, who has always believed in me, and

this has permitted me to be what I am, and therefore to reach this

important goal.

Contents

Abstract ............................................................................................. 1

........................................................................................... 3

Overview of the Thesis ..................................................................... 3

Motivations ........................................................................ 4

Activity and results ............................................................ 6

Summary ............................................................................ 7

Publications in Journal and Conference Papers ................. 8

......................................................................................... 13

Introduction ..................................................................................... 13

Background ...................................................................... 14

Ionic liquids ...................................................................... 15

Refractory metals ............................................................. 18

Refractory metals deposition in high temperature molten

salts .......................................................................................... 20

Refractory metals deposition in ionic liquids................... 24

......................................................................................... 27

Materials and methods .................................................................... 27

Experimental methods ...................................................... 28

3.1.1. Materials and Chemicals ........................................... 28

3.1.2. Electrochemical set-up .............................................. 29

Electrochemistry .............................................................. 30

3.2.1. Cyclic voltammetry ................................................... 30

3.2.2. Pre-electrolysis process ............................................. 31

3.2.3. Electrodeposition process ......................................... 31

Surface and structure analysis .......................................... 32

3.3.1. Scanning electron microscopy ..................................32

3.3.2. X-Ray diffraction ......................................................33

.........................................................................................35

Results and discussion .....................................................................35

Introduction ......................................................................36

4.1.1. Treatment of Ionic Liquid .........................................36

Electrochemical deposition of niobium ............................39

4.2.1. Boron doped diamond substrate ................................39

4.2.2. Polycrystalline gold substrate....................................48

4.2.3. Copper substrate ........................................................51

Electrochemical deposition of tantalum ...........................54



4.3.3. Copper substrate ........................................................61

Electrochemical deposition of zirconium .........................65



Electrochemical deposition of copper ..............................76

4.5.1. Dissolution of pure copper metal ..............................76

4.5.2. Niobium substrate .....................................................77

4.5.3. Tantalum substrate ....................................................82

Electrochemical deposition of composites .......................83

4.6.1. Electrodepostion of Cu/Nb composites .....................84

4.6.2. Electrodepostion of Cu/Ta composites .....................86

.........................................................................................91

Conclusions .....................................................................................91

Appendix .........................................................................................95

Electrochemical processes ............................................... 96

A.1.1. Nonfaradaic processes .............................................. 98

A.1.2. Faradaic processes .................................................. 100

A.1.3. The Butler-Volmer equation ................................... 102

Electrochemical techniques ............................................ 105

A.2.1. Potential sweep techniques and cyclic voltammetry ....

................................................................................ 105

A.2.2. Reversible reactions ................................................ 107

A.2.3. Irreversible systems ................................................ 109

A.2.4. Quasi reversible systems ......................................... 110

Potential step techniques ................................................ 111

Bibliography ................................................................................. 117

Table of symbols ........................................................................... 127

List of figures ................................................................................ 133

Abstract 1

Abstract

Multilayer metal materials with nanometric scale are important in

modern engineering applications. Composite materials based on

metals with different characteristics give rise to new materials with

unique properties. Among the others, nanostructured composites

constituted by immiscible metals can present interfaces able to

control defects produced by high doses of radiation, stress and

temperature: their properties can be exploited in nuclear power

reactor. Immiscible systems constituted by Cu/Nb or Cu/Ta

multilayers exhibit higher thermal stability and improved mechanical

properties with respect to bulk Nb, bulk Ta and bulk Cu.

Refractory metals present high melting point, high hardness and high

resistance against strong acids and bases. The electrodeposition of

these metals presents several limitations: the most important is the

very negative deposition potential that makes difficult the deposition

of metals such as niobium, tantalum and zirconium. Since both

oxygen reduction and hydrogen evolution from water splitting occur

at potential values much less cathodic than the metals reduction, at

the low potential requested to obtain niobium, tantalum and

zirconium in metal form it is necessary the use of electrolytes free of

water and oxygen and characterized by high stability in large

potential windows. To overcome this limitation, the

electrodeposition from molten salts or ionic liquids as solvents have

been proposed.

In the present project the electrochemical coating of niobium,

tantalum zirconium and copper has been investigated in 1-butyl-1-

methylpyrrolidinium bis(trifluoromethylsulfonyl) imide

2 Abstract

([BMP][TFSA]) on both boron doped diamond (BDD) and metal

substrates in order to determine the reduction path for both single

metal and nanometric composites electrodeposition.

Electrochemical experiments have been performed at different

temperatures in a glove box, under nitrogen atmosphere.

Galvanostatic runs and cyclic voltammograms performed at different

scan rates and different potential windows have been carried out in

order to determine the behaviour of the systems employed.

Potentiostatic experiments were performed at the potential values

corresponding to the voltammetric peaks and the samples obtained

were analysed by SEM-EDX analyses.

Regarding the electrodeposition of refractory metals nanometric

crystallites have been obtained at 125 °C.

Cu/Nb and Cu/Ta composites have been prepared by a dual bath

deposition technique; the deposits were constituted by fine

crystallites with average sizes in the range 50-100 nm. The elemental

maps indicate a different distribution of Cu/Nb and Cu/Ta in the

composites obtained with the different substrates.

Overview of the Thesis

This Chapter starts with the illustration of the motivations which lead

us to the development of this Thesis. After that, a summary of the

Thesis is given, by showing how it is structured in different Chapters.

Finally, a list of publications in journals and conference papers and

other activities of the author is presented.

4 Chapter 1. Overview of the Thesis

Motivations

Multilayer metal materials with nanometric size are important in

modern engineering applications. Composite materials based on

metals with different characteristics give rise to new materials with

unique properties. Among the others, nanostructured composites

constituted by immiscible metals can present interfaces able to

control defects produced by high doses of radiation, stress and

temperature: their properties can be exploited in nuclear power

reactors, where materials withstanding elevated temperatures and

radiation fluxes for long time in a corrosive environment are

requested.

Niobium, tantalum and zirconium are so-called refractory metals:

they present high melting point, high hardness and high resistance

against strong acids and bases.

Thin-film composites of Cu/Nb and Cu/Ta exhibit high thermal

stability and mechanical properties; moreover, high radiation

damage-resistance of Cu/Nb composites was observed during

radiation at room and elevated temperatures. This is of great

relevance for materials used in fission and fusion reactors, which may

be submitted to higher doses of radiations and may manifest

significant damages depending on the types of reactions occurring in

the reactor.

Immiscible systems constituted by Cu/Nb or Cu/Ta multilayers

exhibit higher thermal stability and improved mechanical properties

with respect to bulk Nb, bulk Ta and bulk Cu.

The electrochemistry of refractory metals (IV, V and VI groups) has

been widely studied in molten salts to optimize the cathodic

deposition process of these metals. Due to their properties, such as

high strength and high corrosion resistance, refractory metals are

widely used in electronics, optics, sensors, automotive, nuclear and

aerospace industries.

Chapter 1. Overview of the Thesis 5

Nanoscale composites coatings have been obtained by using several

techniques, including electrodeposition, electron-beam evaporation,

and magnetron sputtering.

The electrodeposition of refractory metals presents several

limitations: the most important is the very negative deposition

potential that makes difficult the deposition. Since both oxygen

reduction and hydrogen evolution from water splitting occur at

potential values much less cathodic than the metals reduction, at the

low potential requested to obtain niobium, tantalum and zirconium in

metal form it is necessary the use of electrolytes free of water and

oxygen and characterized by high stability in large potential

windows.

To overcome these limitations, the electrodeposition from molten

salts or ionic liquids as solvents have been proposed. Other than

industrial processes, such as the Hall-Heroult for alluminium, several

case studies are reported in the literature for refractory metals

deposition from molten salts. However, despite they have good ionic

conductivity and wide potential windows, the main disadvantage is

that the metal deposit obtained from these media is dendritic and

sometimes powdery. Furthermore, there are severe issues such as

corrosion problems in processes with high temperatures. Recently,

the use of room temperature ionic liquid (RTIL) has received great

attention due to the low operative temperatures and the high potential

window of stability: these properties make RTIL ideal solvents in the

electrodeposition of refractory metals.

Ionic liquids show good electrical conductivity, high thermal stability

and have wide electrochemical windows; their properties allow the

direct electrodeposition of crystalline metals and semiconductors at

elevated temperatures and, in this context, ionic liquids can be


regarded as the link to molten salts. Ionic liquids give an extension

of the area of application of the coatings of refractory metals.

Activity and results

In the present project the electrochemical deposition of niobium,

tantalum and zirconium has been studied using ionic liquids as

solvent.

The process has been investigated in 1-butyl-1-methylpyrrolidinium

bis(trifluoromethylsulfonyl)imide [BMP][TFSA] on boron doped

diamond (BDD) or metal substrates in order to determine the

reduction path for either single metal or nanometric composites

electrodeposition.

Electrochemical experiments have been performed at different

temperatures in a glove box, under nitrogen atmosphere.

Galvanostatic runs and cyclic voltammograms performed at different

scan rates and different potential windows have been carried out in

order to determine the behaviour of the systems. Potentiostatic

experiments were performed at the potential values corresponding to

the voltammetric peaks and the samples obtained were analysed by

SEM-EDX analyses.

Regarding the electrodeposition of refractory metals, nanometric

crystallites have been obtained at 125 °C.

Cu/Nb and Cu/Ta composites have been prepared by a dual bath

deposition technique; the deposits were constituted by fine

crystallites with average sizes in the range 50-100 nm. The elemental

maps indicate a different distribution of Cu/Nb and Cu/Ta in the

composites obtained with the different substrates.


Summary

This Thesis is structured in five main chapters subdivided in

paragraphs for each topic, in order to allow the reader to better follow

the development of the research project. Electrochemical processes

and techniques are reported in the Appendix.

A summary of this Thesis is shown in the following list, where a brief

description of each chapter is given.

Chapter 2. A general overview of the electrodeposition of refractory

metals both in high temperature molten salts and in low temperature

molten salts (or ionic liquids) is given, focusing the attention on the

properties of ionic liquids. This chapter introduces the reader from

the basics of electrodeposition of refractory metals.

Chapter 3. Here, all the procedures to obtain the experimental data

used to study the processes investigated are explained as well as all

the devices used. The techniques used to characterize the samples

obtained are also explained.

Chapter 4. This Chapter shows all the experimental results obtained

for refractory metals electrodeposion and for the development of

nanocomposites materials, focusing the attention on the use of

different substrates as working electrode.

Chapter 5. In this Chapter conclusions are given.

Appendix. This chapter is focused on the electrochemical processes

taking place at electrode surface and electrochemical techniques used

to investigate the electrochemical behaviour of electrode materials.


Publications in Journal and Conference Papers

Some of the topics present in this Thesis have been published in

international journal papers, national and international

congresses/conferences.

International Journal Papers:

A. Vacca, M. Mascia, L. Mais, S. Rizzardini, F. Delogu, S. Palmas,

On the Electrodeposition of Niobium from 1-Butyl-1-

Methylpyrrolidinium Bis(trifluoromethylsulfonyl)imide at

Conductive Diamond Substrates, Electrocatalysis (2014) 5:16-22

M. Mascia, A. Vacca, L. Mais, S. Palmas, E. Musu, F. Delogu,

Electrochemical deposition of Cu and Nb from pyrrolidinium based

ionic liquid, Thin Solid Films (2014) 571:325-331

Laura Mais, Michele Mascia, Annalisa Vacca, Simonetta Palmas,

Francesco Delogu, Voltammetric Study on the Behaviour of

Refractory Metals in ([BMP][TFSA]) Ionic Liquid, Chemical

Engineering Transactions (2014) 41:97-102

Annalisa Vacca, Michele Mascia, Laura Mais, Francesco Delogu,

Simonetta Palmas, Alessandra Pinna, Electrodeposition of Zirconium

from 1-butyl-1-methylpyrrolidinium-bis(trifluoromethylsulfonyl)

imide: electrochemical behaviour and reduction pathway, Materials

and Manufacturing Processes, DOI 10.1080/

10426914.2015.1004698.

Not yet published International Journal Papers:


Francesco Delogu, Electrochemical deposition of Cu/Ta composites

http://www.sciencedirect.com/science/journal/00406090


from pyrrolidinium based ionic liquid [Submitted to “Journal of

Applied Electrochemistry”]

National and International Congresses and Conferences:

Laura Mais, Michele Mascia, Annalisa Vacca and Simonetta Palmas,

On the Electrodeposition of Niobium and Copper in a Ionic Liquid

for Multilayers Preparation, ELECTROCHEM 2013, Southampton

(UK), September 2013

M. Mascia, A. Vacca, L. Mais, S. Palmas, On the Electroreduction

of Niobium from Ionic Liquid at Different Temperatures, 64th Annual

Meeting of the International Society of Electrochemistry (Mexico),

September 2013

M. Mascia, A. Vacca, L. Mais, S. Palmas, Electrochemical behaviour

of Nb, Ta, Zr, and W in pyrrolidinium based ionic liquid, 64th Annual

Meeting of the International Society of Electrochemistry (Mexico),

September 2013

M. Mascia, A. Vacca, S. Palmas, L. Mais, S. Rizzardini, F. Delogu,

Electrochemical deposition of Cu and Nb in Pyrrolidinium based

Ionic Liquid for Multilayers preparation, Multilayers’13, Madrid

(Spain), October 2013


Simone Rizzardini, Francesco Delogu, On the electroreduction of

Tantalum from 1-Butyl-1-Methylpyrrolidinium

Bis(trifluoromethylsulfonyl)imide, XXXV RGERSEQ-1STE3, Burgos

(Spain), July 2014

L. Mais, M. Mascia, A. Vacca, S. Palmas, F. Delogu,

Electrodeposition of Nb-Cu and Ta-Cu composites from 1-butyl-1-

methylpyrrolidinium bis(trifluoromethylsulfonyl) imide, 65th


Annual Meeting of the International Society of Electrochemistry,

Lausanne (Switzerland), September 2014

L. Mais, M. Mascia, A. Vacca, S. Palmas, F. Delogu, Voltammetric

study on the behaviour of refractory metals in ([BMP][TFSA])

ionic liquid, 10th European Symposium on Electrochemical

Engineering, Chia, Sardinia (Italy), October 2014

In addition to the work presented in this Thesis the author has also

participated, during the course of the study, in other research

projects:

International Journal Papers:

A. Vacca, M. Mascia, S. Palmas, L. Mais, S. Rizzardini, On the

formation of bromate and chlorate ions during electrolysis with

boron doped diamond anode for seawater treatment, Journal of

Chemical Technology and Biotechnology (2013) 88:2244-2251

A.Vacca, M.Mascia, S.Rizzardini, S.Palmas, L.Mais, Coating of gold

substrates with polyaniline through electrografting of aryl diazonium

salts, Electrochimica Acta (2014) 126:81-89

National and International Congresses and Conferences:

M. Mascia, A. Vacca, S. Palmas, A. Da Pozzo, L. Mais, Trattamento

Elettrochimico Per La Rimozione Di Chlorella Vulgaris Con Anodi

Di BDD, GEI-ERA2012, Santa Marina Salina, Messina (Italy), June

2012

http://www.scopus.com/source/sourceInfo.url?sourceId=16083&origin=recordpage

http://www.scopus.com/source/sourceInfo.url?sourceId=16083&origin=recordpage


S. Palmas, A. Da Pozzo, M. Mascia, A. Vacca, L. Mais, C. Ricci,

Sensibilizzazione Di Nanostrutture Di TiO2 Con Cumarina, GEI-

ERA2012, Santa Marina Salina, Messina (Italy), June 2012

A. Lallai, L.Mais, Ruolo del Triton X100 nella biodegradazione del

fenantrene, GRICU 2012, Montesilvano (Italy), September 2012

M. Mascia, A. Vacca, S. Palmas, A. Da Pozzo, L. Mais, Rimozione

di Microalghe Mediante Trattamento Elettrochimico con Anodi di

Diamante Conducente, GRICU 2012, Montesilvano (Italy),

September 2012

A. Vacca, M. Mascia, S. Palmas, L. Mais, S. Rizzardini,

Electrochemical preparation of Gold/Polyaniline electrodes through

electrografting of diazonium Salts, GEI2013, Pavia (Italy),

September 2013

M. Mascia, A. Vacca, S. Palmas, L. Mais, Three Dimensional

Electrodes For The Removal Of Microalgae From Water, 9th

European Congress of Chemical Engineering, The Hague (The

Netherlands), April 2013

Simonetta Palmas, Annalisa Vacca, Michele Mascia, Simone

Rizzardini, Laura Mais, Elodia Musu, Esperanza Mena and Javier

Llanos, Synthesis and characterization of ordered TiO2

nanostructures, ELECTROCHEM 2013, Southampton (UK),

September 2013

P.Ampudia, S.Palmas, A.Vacca, M.Mascia, L.Mais, S.Monasterio,

S.Rizzardini, E.Musu, Systematic investigation on the effect of the

synthesis conditions on the performance of nanotubular structured

electrodes, SiO2 Advanced dielectrics and related devices, X

International Symposium, Cagliari (Italy), June 2014

S. Monasterio, F. Dessì, M. Mascia, A. Vacca, S. Palmas, L. Mais,

Electrochemical treatment for the removal of Chlorella Vulgaris

http://www.aidic.it/gricu/Leaflet%20Convegno%20Scuola%20GRICU.pdf

http://www.aidic.it/gricu/Leaflet%20Convegno%20Scuola%20GRICU.pdf


and Microcystis aeruginosa by using a fixed bed single cell, 65th

Annual Meeting of the International Society of Electrochemistry,

Lausanne (Switzerland), September 2014

L. Mais, M.J. Martín de Vidales, C. Ponce de León Albarran,

Photoelectrocatalytic oxidation of Methyl Orange with highly

ordered TiO2 nanotubes array, 10th European Symposium on

Electrochemical Engineering, Chia, Sardinia (Italy), October 2014

Introduction

A general overview of the electrodeposition of refractory metals both

in high temperature molten salts and in low temperature molten salts

(or ionic liquids) is given, focusing the attention on the properties of

ionic liquids. This chapter introduces the reader from the basics of

electrodeposition of refractory metals

14 Chapter 2. Introduction

Background

Electrodeposition of metals is essential for a variety of industries

including electronics, optics, sensors, automotive and aerospace. Due

to their physical proprieties such as superconductivity, thermal and

mechanical stability, refractory metals are widely used in several

application fields. The mechanical properties of refractory metals,

such as the excellent corrosion resistance and the nuclear properties

of some, make them particular interesting as coatings.

Electrodeposition of metals from a liquid bath containing metal salts

is the state of the art in industrial coatings: the technique has shown

to be affordable and economically sustainable for very large coatings.

Nanoscale composites coatings have been obtained by using several

techniques, including electrodeposition [1], electron-beam

evaporation, severe plastic deformation [2] and magnetron sputtering

[3].

Thin-film composites of Cu/Nb and Cu/Ta exhibit higher thermal

stability and mechanical properties with respect to bulk Cu, bulk Nb

or bulk Ta. Moreover, high radiation damage-resistance of Cu/Nb

composites was observed during radiation at room and elevated

temperatures [4] [5].

This is of great relevance for materials used in fission and fusion

reactors, which may be submitted to higher doses of radiation and

may manifest significant damage depending on the types of reactions

occurring in the reactor.

The electrochemical deposition of Cu is currently achieved from

aqueous solutions, while the deposition of refractory metals from

aqueous media is virtually impossible, due to the wide negative

deposition potential. High temperature molten salts are often

proposed in literature for electrodeposition of refractory metals

although these baths have many technical and economic problems,

Chapter 2. Introduction 15

such as the loss in the current efficiency, due to the dissolution of

metal after its deposition, and corrosion problems at high

temperatures. Recently, the use of room temperature ionic liquid

(RTIL) has received great attention due to the lower operative

temperatures and the high potential window of stability. These

properties make RTIL ideal solvents in the electrodeposition of

metals.

Ionic liquids

Since ionic liquids have extraordinary physical properties superior to

those of water or organic solvents, they have attracted considerable

attention in recent years.

The ionic liquids based on AlCl3 can be regarded as the first

generation of ionic liquids. In 1970s and 1980s, Osteryoung et al. [6]

[7] and Hussey et al. [8] [9] carried out extensive research on organic

chloride–aluminium chloride ambient temperature ionic liquids and

the first major review of room temperature ionic liquids was written

by Hussey [10].

The synthesis of air and water stable ionic liquids, which are

considered as the second generation of ionic liquids, attracted further

interest in the use of ionic liquids in various fields. In 1992, Wilkes

and Zaworotko [11] reported the first air and moisture stable ionic

liquids based on 1-ethyl-3-methylimidazolium cation with either

tetrafluoroborate or hexafluorophosphate as anions. Unlike the

chloroaluminate ionic liquids, these ionic liquids could be prepared

and safely stored outside of an inert atmosphere. Generally, these

ionic liquids are water insensitive, however, the exposure to moisture

for a long time can cause some changes in their physical and chemical

properties. Ionic liquids based on more hydrophobic anions such as

bis-(trifluoromethanesulfonyl) imide and tris-


(trifluoromethanesulfonyl) methide have received extensive attention

not only because of their low reactivity with water but also because

of their large electrochemical windows [12].

Usually these ionic liquids can be well dried the water contents below

1 ppm under vacuum at temperatures between 100 and 150°C.

Ionic liquids are basically constituted of ions and behave very

differently than conventional molecular liquids when they are used

as solvents. In contrast to conventional molecular solvents, ionic

liquids are usually nonvolatile, non-flammable and less toxic; are

good solvents for both organics and inorganics and can be used over

a wide temperature range, in fact ionic liquids can be thermally stable

up to temperatures of 450 °C [13].

Ionic liquids have reasonably good ionic conductivities compared

with those of organic solvents/electrolyte systems (up to ~ 10 mS

cm-1); however, at room temperature their conductivities are usually

lower than those of concentrated aqueous electrolytes. The

conductivity of ionic liquids is inversely linked to their viscosity.

Hence, ionic liquids of higher viscosity exhibit lower conductivity.

Increasing the temperature increases conductivity and lowers

viscosity.

Ionic liquids have been defined to have melting points below 100 °C

and most of them are liquid at room temperature. Both cations and

anions contribute to the low melting points of ionic liquids. The

increase in anion size leads to a decrease in melting point. Cations

size and symmetry make an important impact on the melting points

of ionic liquids. Large cations and increased asymmetric substitution

result in a melting point reduction [14].

Moreover, ionic liquids can have pretty large electrochemical

windows, more than 5 V, and hence they give access to elements that

cannot be electrodeposited from aqueous or organic solutions.


The electrochemical window is an important property and plays a key

role in using ionic liquids in electrodeposition of metals and

semiconductors. By definition, the electrochemical window is the

electrochemical potential range over which the electrolyte is neither

reduced nor oxidized at an electrode. This value determines the

electrochemical stability of solvents. As known, the

electrodeposition of elements and compounds in water is limited by

its low electrochemical window of only about 1.2 V.

In general, the wide electrochemical windows of ionic liquids have

opened the door to electrodeposit metals and semiconductors at room

temperature which were formerly obtained only from high

temperature molten salts. The thermal stability of ionic liquids allows

to electrodeposit Ta, Nb, V, Se and presumably many other ones at

elevated temperature [14].

Another advantage of ionic liquids is that problems associated with

hydrogen ions in conventional protic solvents can be eliminated in

ionic liquids because most of the commercially available ionic

liquids are aprotic. Therefore ionic liquids, especially air and water

stable ones, are considered promising solvents for a wide variety of

applications including electrodeposition, batteries, catalysis,

separations, and organic synthesis [15].

Nevertheless, they can be used efficiently for the coating of other

metals with thin layers of tantalum or aluminium.

In several recent reviews ionic liquids are recommended for

electrodeposition [14] [16] [17] [18].

For ionic liquids to be used efficiently, they need not only be

economically viable but also simple to handle and recyclable. While

it is impractical that ionic liquids will compete with all aqueous

solutions in large markets where the current technology works

effectively, it is realistic that they should proffer an alternative for the

deposition of metals that can only currently be applied using vapour

deposition [18].


Refractory metals

The elements of groups IV, V and VI of the periodic table are called

refractory metals due to their high melting points and the refractory

nature of their compounds. There are twelve metals with melting

points equal to or greater than that of chromium, a temperature which

is commonly accepted as the lower melting point limit for defining a

refractory metal. Mechanical properties of metals are structure-

sensitive; this fact is related to the various types of defects which

exist in the crystalline lattice of metals.

They are known for their high melting point, high temperature

strength, irradiation damage resistance and excellent corrosion

resistance which make them ideal metals for use in the chemical and

nuclear industry [19]. In addition, these materials exhibit remarkable

resistance to the harshest of chemical environments due to the

presence of a tenacious oxide film that forms spontaneously in air.

The nuclear and aerospace industries need materials which can

withstand extreme conditions, so that the demand of refractory metals

have increased in recent years.

Niobium have high melting point, lower density and low thermal

neutron cross-section compared to other refractory metals, which

makes niobium useful in atomic reactors. Niobium is highly

recommended in nuclear reactors because of its good high

temperature creep strength, its resistance to corrosion by liquid

sodium-potassium alloys and its outstanding compatibility with

nuclear fuels. However, because of the poor oxidation resistance of

high temperatures of niobium-base alloys, they must be coated before

use in nuclear environments. The corrosion resistance of niobium is

similar to, but not quite as good as that of tantalum, being less

resistant to very strong acids.

Tantalum has excellent corrosion resistance; this property make it an

ideal coating material for components exposed to high temperature


and severe chemical environments [20]. In the nuclear power

industry, tantalum is applied in condensers, columns, heat

exchangers, helical coils, valve linings and a variety of other

components exposed to extremely corrosive fluids [21]. Tantalum

thin films exhibit two crystalline phases, body-centered cubic (alpha

phase, the bulk structure of tantalum) and a meta-stable tetragonal

beta-phase. The structure of deposited thin films is usually the

metastable beta-phase or a mixture of the two phases [20].

It has been found that when depositing high quality tantalum films

and particularly thicker coatings using molten salts [22], the beta-

phase dominates the deposits on most substrates under most

processing conditions [23]. The metastable beta-phase converts to the

alpha-phase upon heating to 750 ˚C, but annealing of the deposited

film to this temperature would adversely affect the properties of a

heat treated substrate [24]. For this reason, deposition of tantalum on

substrates at high temperatures is not suitable. Therefore, it is

preferred to electroplate refractory metals, such as tantalum using

ionic liquids, a process that can be carried out at significantly lower

temperatures compared to other plating processes.

Zirconium metal has a low absorption cross-section for thermal

neutrons, which makes it ideal for nuclear energy uses; this metal has

strong corrosion-resistance properties as well as the ability to confine

fission fragments and neutrons so that thermal or slow neutrons are

not absorbed and wasted, thus improving the efficiency of the nuclear

reactor. The use of zirconium in the construction of nuclear fuel

elements and other structural components in reactor cores meets

many physical and technological requirements essential for the

nuclear industry [25].


Refractory metals deposition in high temperature

molten salts

The electrolysis of molten salts permits pure, plastic, nonporous, and

coherent layers of refractory metals to be coated on substrates of

complex configuration and large size.

The mechanism of the electrodeposition of refractory metals is not

yet fully elucidated. Senderoff and Mellors [26] developed a general

process for the electrodeposition of eight of the nine refractory metals

of groups IV, V and VI as dense coherent deposits by using a solution

of the refractory metal in a molten alkali-fluoride eutectic mixture

from 680 °C to 800 °C. They concluded that an irreversible metal-

producing step is a necessary (though not sufficient) condition for the

deposition of coherent deposits from molten salts.

Inman and White [27] have summarized in a review the

electrochemistry of refractory metals in molten salts electrolytes. The

authors have reported that the recovery of a primary metal is ensured

by using, as solvent, a melt having a much larger decomposition

voltage than that of the solute and an inert cathode; controlling the

electrodeposition of metal by rate processes other than mass transfer,

will lead to coherent deposits.

Few years later, Girginov et al. [28] have presented a selective review

of the anodic and cathodic processes of Ti, Zr, Nb and Ta from

molten salts-based electrolytes in order to investigate the

optimization of cathodic deposition of these metals highlighting that

the two main types of electrolyte employed in the electrodeposition

of refractory metals were chloride and fluoride based.

Despite molten salts have large electrochemical windows, good heat

capacity, can attain very high temperatures and conduct electricity,

the extreme operating conditions limit the variety of substrates used


for the refractory metals electrodeposition and allow to such technical

and economical problems as the loss in the current efficiency and

corrosion problems at high temperatures. Thus, the electrodeposition

of refractory metals from ionic liquid at low temperatures represents

a valid alternative to the use of high-temperature molten salts.

The first result on the electrodeposition of coherent, dense deposits

of refractory metals using mixtures of alkali metal fluorides have

been reported by Mellors and Senderoff [29]. To obtain deposits of

niobium they used eutectic composition of KF-LiF and KF-NaF as a

solvent and NbF5 as a source of Nb at temperature of about 775 °C

obtaining columnar structure of electrodeposits.

The metallic Nb obtained from electrodeposition processes carried

out by employing molten LiF-KF-K2NbF7 baths at temperatures

between 650 °C and 850 °C exhibited the necessary purity for its use

in the preparation of superconducting tapes, in the nuclear technology

and in metallurgy [30].

The influence of temperature on Nb plating using LiCl-KCl and

NaCl-KCl melts was studied by Gillersberg et al: at temperatures

below 550 °C no coherent layer was obtained, at temperatures

between 550 °C and 650 °C the Nb layers were highly heterogeneous,

and characterized by many undesired inclusions. Finally, at a

temperature of 750 °C homogeneous layers of metallic Nb were

deposited [31].

The Ta(V) reduction has been studied in a wide range of melts, such

as chlorine or fluorine melts.

In FLINAK-K2TaF7 Polyakova et al. [32] have studied the nature of

the tantalum reduction process at 710 °C using various working

electrodes, such as molybdenum, platinum and silver; the authors

affirm that the cyclic voltammograms shape were not reproducible


and electrochemically TaF72- is reduced to tantalum metal in a single

quasi-reversible five-electron step.

In LiF-NaF-K2TaF7 with low oxygen contents on Ag electrode,

Chamelot et al. [33] recorded a single peak in the reduction sense

associated with one reoxidation peak on the reverse scan. The

reduction peak reveals the one-step reaction involving pentavalent

and zerovalen tantalum at 800 °C.

In chloride electrolytes [28] suggest that the electrodeposition of

microcrystalline Ta took place in a narrow potential range in chloride

melts and confirmed a single step mechanism in NaCl-KCl-K2TaF7

melts.

The effect of fluoride ions on the reduction of zirconium has been

studied by Guang-Sen et al. [34] in fluoride-chloride melts: the

voltammetric behaviour of zirconium at 973 K and 1033 K in

equimolar KCl-NaCl with ZrCl4 on a platinum electrode showed that

the electroreduction process of Zr is reversible. According to the

number of electrons exchanged, the authors established that the

process has double two-electron steps: Zr(IV)/Zr(II) and Zr(II)/Zr.

The cathodic reduction of Zr(IV), in the same electrolytic bath, was

also studied with different concentrations of KF: a new peak

appeared when small amount of KF (XF-/XZr

4+= 1.07) were added

into equimolar KCl-NaCl melts with ZrCl4. The currents at peaks

related to Zr(IV)/Zr(II) and Zr(II)/Zr reduction decreased as the ratio

(XF-/XZr

4+) in the melts increased. The authors have concluded that

the new peak recorded in the presence of KF represents the main

reduction of Zr when the ratio XF-/XZr

4+ ≥ 6.

Kawase and Ito [35] have studied the voltammetric behaviour on Ni

electrode in the solution obtained after the anodic dissolution of Zr in

the LiCl-KCl eutectic. Cyclic voltammograms reveal that Zr(II) and

Zr(IV) can exist in molten salts between 450 and 550 °C and the ratio

of Zr(II)/Zr(IV) increases with increasing temperature. At 450 °C


only one cathodic peak related to Zr(IV)/Zr(II) is observed, while at

550 °C two cathodic peaks, related to the reduction of Zr(IV)/Zr(II)

and Zr(II)/Zr, are observed.

The electroreduction process of Zr(IV) was studied by Chen et al.

[36] at molybdenum electrode in LiCl-KCl-K2ZrF6 molten salts at

923 K; the authors highlighted that Zr(IV) was reduced to Zr metal

by a two-step mechanism corresponding to the Zr(IV)/Zr(II) and

Zr(II)/Zr transitions. Furthermore, the cathodic peaks are not

dependent on the sweep rate so it has been concluded that the

reduction process of Zr(II)/Zr and Zr(IV)/Zr(II) are both reversible

(or quasi-reversible).

Analogous behaviour has been reported by Wu et al. [37] that studied

the electrochemical reduction of Zr(IV) on Pt electrode at 1023 K in

NaCl-KCl-K2ZrF6 molten salt; two cathodic reduction peaks related

to Zr(IV)/Zr(II) and Zr(II)/Zr steps were observed in the potential

range from 0 to –1.6 V vs. Ag/AgCl. The authors thus observed a

two-electron transfer process involving the formation of Zr(II).

On the other hand, Groult et al. [38] studied the electrochemical

behaviour of Zr(IV) in potassium-free fused alkali fluorides in the

temperature range of 773-1123 K by using either tungsten or

molybdenum electrodes, and observed a single well-defined oxido-

reduction wave, and concluded that the reduction of Zr(IV) is a single

four-electron reversible step controlled by diffusion process.

A detailed electrochemical study of the molten LiF-CaF2-ZrF4

system is provided in the 810-920 °C temperature range by Gibilaro

et al. [39] on an inert Ta electrode. The authors have reported that at

840 °C a single peak is observed in the cathodic run associated with

a stripping peak; the reduction mechanism was a one step process

exchanging four electrons and controlled by the Zr ions diffusion in

the molten salt.


Refractory metals deposition in ionic liquids

Few data on the electrochemistry of refractory metals in ionic liquids

can be found in literature.

The electrodeposition of Nb from low temperature ionic liquids was

investigated by Babushkina et al. [40]; the authors obtained a

relatively detailed picture of the electrochemical behaviour of NbCl5

in 1-butyl-1-methylpyrrolidinium chloride by using electrochemical

studies and Raman spectroscopic measurements. Several similarities

were found with the behaviour obtained for fused alkali metal

solvents, so that the authors concluded that high temperature molten

salts and ionic liquids may be considered as “two sides of the same

coin”. However, metallic Nb deposits were not obtained under the

adopted conditions.

Zein El Abedin et al. [41] have shown the differences on the

electrochemical behaviour of [BMP][TFSA] ionic liquid with

different concentrations of TaF5 either in presence or absence of LiF

on platinum electrode; cyclic voltammograms clearly indicate that

three reduction peaks appear in the presence of LiF while only two

reduction peaks, correlated to the electrolytic reduction of Ta(V) to

Ta(III) and to the reduction of Ta(III) to Ta(0), have been recorded

without LiF addition.

The two-step reduction mechanism of Ta(V) was confirmed few

years later by Borisenko et al. [21] that studied the electrochemical

behaviour of TaF5 in [Py1,4]TFSA on Au(111) at 25 °C.

The electrochemical behaviour of Ta(V) chloride and oxochloride

species has been also studied by Babushkina and Ekres [42] in

Pyr14Cl ionic liquid: the authors reported that the mechanism of the

electrochemical reduction of Ta(V) depends strongly on the

composition of the electrolytic bath.


Few data on the electrochemistry of Zr(IV) using room temperature

ionic liquids as solvent has been hitherto reported in the literature.

Tsuda et al. obtained Al-Zr alloys working with aluminium chloride-

1-ethyl-3-methylimidazolium chloride molten salt at 353 K [43];

however, the reduction process was complicated by the formation of

insoluble ZrCl3.

More recently Compton et al. studied the electrochemistry of ZrCl4

in N-Butyl-N-methylpyrrolidinium- bis(trifluoromethylsulfonyl)

imide at 25°C [44]: as in the case of chloroaluminate molten salts,

the reduction of zirconium in pyrrolidinium-based ionic liquid was

complicated by the formation of Zr(III). Moreover, the authors

highlighted the interesting use of ZrCl4 as dry agent for ionic liquid

due to the formation of insoluble ZrCl2O by hydrolysis of zirconium

chloride in the presence of water in the solvent.

Materials and methods

Here, all the procedures to obtain the experimental data used to study

the processes investigated is explained as well as all the devices used.

The techniques used to characterize the samples obtained are also

explained.

28 Chapter 3. Materials and methods

Experimental methods

3.1.1. Materials and Chemicals

The air and water stable ionic liquid 1-butyl-1-methylpyrrolidinium

bis(trifluoromethylsulfonyl)imide [BMP][TFSA], NaF (99.99 %),

LiF (99.99 %), NbF5 (98 %), TaF5 (98 %) and ZrF4 (99.9 %) were

purchased from Aldrich.

The purity of the ionic liquid given by the supplier is ≥ 98.5 %. The

presence of impurities in commercial ionic liquids (even in ultrapure

quality) can alter the deposition process in initial stages, leading to

misinterpretation of the surface processes [21]. Prior to use, ionic

liquid was dried at 125 °C for 24 h under stirring in order to remove

water content and stored in a nitrogen filled glove box (Sicco).

Different substrates, such as silicon coated with boron doped

diamond (Si/BDD, electrode area 0.7 cm2) purchased from Neocoat

(Switzerland), niobium foils (thickness 0.25 mm, 99.8%, Aldrich),

tantalum foils (thickness 0.25 mm, 99.9%, Aldrich), polycrystalline

gold foils (thickness 0.25 mm, 99.9%, Aldrich) and copper foils

(thickness 0.25 mm, 99.98%, Aldrich) were used as working

electrode materials (electrode area 0.6 cm2). The high

electrochemical stability of diamond electrodes, compared to

conventional electrode materials, under several conditions makes

these materials of particular interest. Boron doped diamond

electrodes have some other advantages such as high electrochemical

reproducibility and high thermal stability.

Prior to use metallic working electrodes were cleaned for 10 minutes

in an ultrasonic bath in acetone. A platinum sheet and a platinum wire

were used as counter and quasi reference-electrodes, respectively.

From an electrochemical point of view a Pt quasi-reference electrode

is definitely not a perfect one, but it has the advantage that avoids all

type of contamination.

Chapter 3. Materials and methods 29

3.1.2. Electrochemical set-up

All electrochemical experiments were performed at fixed

temperature ranging from R.T. to 200 °C in the nitrogen filled glove

box as shown in Figure 1.

Figure 1: Nitrogen filled glove box in which all experiments were

performed.

The typical three electrode cell consisted of a pyrex cylindrical

beaker equipped with a counter electrode with an area of 0.8 cm2

immersed in solution, a quasi-reference electrode with an area of 0.15

cm2 immersed in solution and a working electrode (surface area

dependent on the substrate used).

Figure 2: The three electrode cell used for all electrochemical experiments.

The solutions were prepared by adding either NaF or LiF to the ionic

liquid up to a concentration of 0.25 M and stirred until dissolved


maintaining the resulting electrolytic bath at 125 °C for 24 h. The

solution containing either sodium fluoride or lithium fluoride salts

was characterized by cyclic voltammetries before the addition of

metal salts. NbF5, TaF5 and ZrF4 were dissolved in a concentration

of 0.25 M and the electrolytic bath was maintained at 125 °C for 24

h under stirring and again characterized by cyclic voltammetries. It

was observed that the metal salts were not completely dissolved in

the liquid and a partial residue of the salt was seen at the bottom of

the solution. Copper ions were introduced in [BMP][TFSA] ionic

liquid containing LiF (0.25 M) by oxidation of a copper sheet

(Aldrich, 99.98%) in a divided cell a fixed potential of 0.1 V vs. Pt.

During the copper dissolution the solution was maintained at 125°C

and magnetically stirred.

Electrochemistry

3.2.1. Cyclic voltammetry

Electrochemical measurements were carried out using a Metrohm

Autolab 302N potentiostat-galvanostat controlled by NOVA

software (Metrohm, Switzerland).

Cyclic voltammetries were started from the open circuit potential

(OCP) in the negative direction at different potential windows,

different scan rates and different temperatures. In order to avoid ion

concentration gradients in the electrolyte, the ionic solution was

magnetically stirred prior to each scan.


Figure 3: Metrohm Autolab 302N potentiostat-galvanostat controlled by

NOVA software (Metrohm, Switzerland).

3.2.2. Pre-electrolysis process

Cyclic voltammetry was performed on the dried ionic liquid to study

its electrochemical behaviour. In order to reduce residual water active

in solution, potentiostatic experiments were performed by using a

titanium grid coated with platinum electrode fixing the potential at -

2 V for 7000 s. The ionic liquid treatment was conducted inside the

glove box under nitrogen atmosphere. A cyclic scan was then

performed to compare the electrochemical behaviour before and after

the potential hold in the liquid.

3.2.3. Electrodeposition process

Upon purification of the ionic liquid, indicated by extremely low

currents in the i-t curve, either LiF (0.25 M) or NaF (0.25 M) were

added to the ionic liquid and stirred until dissolved at 125 ˚C. Then

0.25 M of metal salt was added into the solution and stirred until

mostly dissolved. The solution was then left in the glove box

overnight to settle and allow any further dissolution of the powders

to occur.

Potentiostatic experiments were performed at different potential

values, determined in the cathodic scan of the voltammogram, to test

the deposition conditions; the potential was kept at fixed value under

stirring. Potentiostatic scans were repeated at different potentials


until deposition was observed. After the electrodeposition tests the

samples were washed in isopropanol and boiled in distilled water for

15 minutes to remove the residual ionic liquid.

Surface and structure analysis

Different analytical techniques, such as scanning electron

microscopy, energy dispersive X-ray analysis and X-ray diffraction

were used to analyse the electrodeposited layers on the substrate.

3.3.1. Scanning electron microscopy

A high resolution Scanning Electron Microscopy (SEM) equipped

with Energy Dispersive X-Ray (EDX) detector (Zeiss, Germany) was

employed to investigate the surface morphology of the deposited

films and to obtain information on the elements present on the

deposited layers.

Figure 4: Scanning Electron Microscopy (SEM) equipped with Energy

Dispersive X-Ray (EDX) detector (Zeiss, Germany).


The EDX analyses were performed usually for every sample at four

magnifications. Point analyses were also performed for some

representative layers. Focused ion beam (FIB) was used to very

precise cross-sections of a sample for subsequent imaging via SEM.

3.3.2. X-Ray diffraction

XRD patterns were collected with a Miniflex II Rigaku

diffractometer, equipped with Cu Kα radiation, over a range of

scattering angle 2θ from 20° to 70°. Discrete angular displacements

of 0.01° and acquisition times of 1 min were used. The XRD patterns

were analysed according to the Rietveld method.

The best fitting of the experimental XRD profiles with a set of

mathematical functions provided information on the relative amounts

of the deposited phases, as well as on their microstructure [45]. The

average size of the coherent diffraction domains and their average

total strain content were estimated under the assumption of isotropic

size and strain content [45].

Results and discussion

This Chapter shows all the experimental results obtained for

refractory metals electrodeposition and for the development of

nanocomposites materials, focusing the attention on the use of

different substrates as working electrode.

36 Chapter 4. Results and discussion

Introduction

Electrochemical conditions and purity of ionic liquid are key factors

in controlling the electrodeposition of refractory metals. The pre-

electrolysis treatment implemented in order to remove the water

impurities from the liquid is discussed below. Than the

electrochemical conditions investigated for electrodeposition of

metals and the microstructure analysis of the electrodeposited layers

are presented.

4.1.1. Treatment of Ionic Liquid

Figure 5: Cyclic voltammetries of pure ionic liquid recorded on BDD

electrode before and after heat treatment at 125°C (scan rate 100 mV/s).

Impurities in ionic liquids can strongly alter the deposition process in

initial stages leading to the misinterpretation of the surface processes.

Consequently, the as-received low-purity ionic liquid, 1-butyl-1-

methyl-pyrrolidinium bis(trifluoromethylsulfonyl)imide was

-10

-8

-6

-4

-2

0

2

4

6

-3.2 -2.2 -1.2 -0.2 0.8 1.8

i /

mA

cm

-2

Potential / V vs Pt-quasi ref.

IL

IL after 24h

Chapter 4. Results and discussion 37

purified before to start the investigation of the behaviour of the

system.

Figure 5 shows the voltammetric behaviour of pure ionic liquid at

125 °C, as-received and after 24 h at 125 °C; BDD was used as

cathode. The ionic liquid exhibits a wide electrochemical window of

stability (4.5 V) where the degradation of the anion and cation

constituting the ionic liquid does not occur.

The anodic peak centred at about 0.5 V could be related to the

presence of impurities: halides or monovalent ions (K+, Na+) are

often detected in commercial ionic liquids, even of ultrapure quality

[21]; moreover, it is difficult to completely avoid contamination of

moisture and oxygen. By extending the potential range from -3.2 to

2.5 V an increase in the current density was observed in the cathodic

and anodic limits probably due to the irreversible reduction of the

[BMP] and the [TFSA] oxidation: under these extreme conditions the

original white colour of the ionic liquid changed to dark-brown.

Figure 6: Decreasing trend of end current densities during the electrolysis

process indicating that impurities are being extracted from the liquid.

-0.45

-0.4

-0.35

-0.3

-0.25

-0.2

-0.15

-0.1

-0.05

0

0 2000 4000 6000 8000

i /

mA

cm

-2

Time / s


Since one of the known impurities was water, the as-received ionic

liquid was dried inside the glove box under nitrogen atmosphere to

remove as much water as possible from the ionic liquid. The

corresponding curve after drying the ionic liquid, submitted to an

electrolysis process for 7000 s, is shown in Figure 6.

Theoretically pure ionic liquid should have no current running

through it, resulting in a linear curve as it is free of impurities.

However, from Figure 7 it is possible to observe that the ionic liquid

purchased for this present study has an end current density of

approximately 0.8 mA·cm-2, indicating that the ionic liquid is not

pure, containing some impurities. A significant increase in current

density can be seen through the ionic liquid, which is probably due

to water reduction or impurities. The exact impurities in the ionic

liquid are unknown.

Figure 7: Cyclic voltammetry before and after pre-electrolysis process

(scan rate 100 mV/s).

-1.2

-0.8

-0.4

0.0

0.4

0.8

-4 -2 0 2 4

i /

mA

cm

-2


pre - electrolysis post - electrolysis


Cyclic voltammetries were recorded before and after the

electrochemical treatment to determine the purity level of the as-

received ionic liquid.

Figure 7 shows that the current density decreases after the electrolysis

process highlighting the reduction of residual water active in

solution. It is evident that there is some current flowing through the

ionic liquid, however not as high as the current observed in the as-

received ionic liquid. This suggests that the water was not entirely

removed and the difficulty of drying the liquid.

Electrochemical deposition of niobium

After purification, either NaF or LiF were added to the IL up to a

concentration of 0.25 M.

Using high temperature molten salts as electrolyte it was found that

the addition of fluorides of alkali melts, such as NaF and LiF to the

electrolytic bath facilitates the electrodeposition of refractory metals

and improves the quality of the deposit obtained [41].

4.2.1. Boron doped diamond substrate

Figure 8 shows the cyclic voltammograms at different temperatures

of [BMP][TFSA] containing 0.25 M LiF and 0.25 M NbF5 at BDD

electrode at a scan rate of 100 mV/s. The electrode potential was

swept initially from the OCP to more negative potentials. The

voltammograms exhibit a different behaviour with increased

temperature: the peak potentials of the cathodic peaks slightly move

to less negative values. The peak currents of the cathodic and anodic

branch significantly increase with rising temperature; the mobility of


the electroactive species towards the electrode surface increases

leading to increasing the reduction and oxidation reaction rate [41].

Figure 8: Cyclic voltammograms of 0.25 M NbF5 and 0.25 M LiF in

[BMP][TFSA] ionic liquid at different temperatures (scan rate 100 mV/s).

An increase of the temperature above 150 °C leads to significant

change of the kinetics of the electrochemical processes, but can also

provoke the destruction of the structure of ionic liquid [42].

Subsequent experimental tests have been conducted at 125 °C: this is

a good compromise to have high conductivity and solubility of metal

salts and to avoid extreme operating conditions.

Figure 9 shows the cyclic voltammograms of ionic liquid containing

0.25 M NaF and 0.25 M NbF5 at BDD electrode at different scan

rates. The electrode potential was scanned from the open circuit

potential in the negative direction down to -2.8 V while the anodic

scan was up to 1.5 V. Three reduction waves can be observed in the

cathodic scan at -1.1 V, -1.7 V and -2.2 V, while in the backward

scan two less defined oxidation waves appear at -1.6 V and 0.5 V.

-8

-6

-4

-2

0

2

4

-3 -2 -1 0 1 2

i /

mA

cm

-2


100 °C

125 °C

150 °C


Figure 9: Cyclic voltammograms at different scan rates of [BMP][TFSA]

containing 0.25 M NbF5 and 0.25 M NaF at BDD electrode at 125 °C.

Figure 10: Cyclic voltammograms at different scan rates of [BMP][TFSA]

containing 0.25 M NbF5 and 0.25 M LiF at BDD electrode at 125 °C.

-4

-3

-2

-1

0

1

2

-3.2 -2.2 -1.2 -0.2 0.8 1.8

i /

mA

cm

-2


25 mV/s

50 mV/s

100 mV/s

200 mV/s

-8

-6

-4

-2

0

2

4

-3.2 -2.2 -1.2 -0.2 0.8 1.8

i /

mA

cm

-2


25 mV/s

50 mV/s

100 mV/s

200 mV/s


Analogous features were observed in the cathodic scan by using LiF

instead of NaF (see Figure 10): the first and the second cathodic

peaks are shifted to -0.68 V and -1.3 V respectively, while the third

cathodic peak remains centred at -2.2 V. In the reverse scan two not

well resolved peaks are present, which can be better analysed by

decreasing the cathodic stop potential in the cyclic voltammetries.

As can be seen from Figure 11, if the reduction scan is stopped at

-2.3 V the oxidation peak at -1.9 V is still present while stopping the

cathodic scan at -2.0 V, it does not appear in the voltammogram but

only the oxidation peak at around -1.1 V is observed.

Figure 11: Cyclic voltammograms of [BMP][TFSA] containing 0.25 M

NbF5 and 0.25 M LiF at BDD electrode at 125° C obtained at different stop

potentials (scan rate 200 mV/s).

A three steps mechanism can be than proposed for Nb(V)

electroreduction.

The Nb(V) reduction has been widely studied in molten salts with

different eutectic compositions. Kuznetsov [46] evidenced a three

-8

-6

-4

-2

0

2

4

-3.2 -2.2 -1.2 -0.2 0.8 1.8

i /

mA

cm

-2



step reduction involving Nb(V), Nb(IV) and Nb(II) at temperature

below 630 °C using LiCl-KCl-NbCl5 as electrolytic bath: the

formation of Nb(II) allowed to the formation of insoluble subhalide

cluster compounds which affected the metal deposition. Different

composition of the melt and temperature of the bath can lead to metal

deposition: the formation of Nb(II) was suppressed by working at

temperature higher than 650 °C, while introducing fluoride ions had

a beneficial effect on the quality of the Nb deposit and stabilized the

higher oxidation states.

A two-step mechanism Nb(V) <=> Nb(IV) <=> Nb(0) was proposed

in NaCl-KCl-K2NbF7 mixture along with the formation of a thin film

of Nb carbide (NbC and Nb2C) when carbon based electrodes were

used [47]. Nb has a strong affinity for carbon, which acts as a

depolarizer for the reduction reaction, so that the formation of a first

layer of carbides with thickness of few atoms cannot be avoided [45].

Babushkina et al. studied the voltammetric behaviour of Nb(V) in 1-

butyl-1-methyl-pyrrolidinium chloride at low temperatures with

Raman spectroscopy: although metallic Nb has not been obtained,

the authors demonstrated many similarities with the behaviour

obtained in high temperature molten salts. In particular, the different

steps of reduction involving Nb(V), Nb(IV) and Nb(III) chloride

complexes were analogous to these observed during Nb reduction in

chloride molten salts [40].

In order to individuate the reduction steps and the possible

mechanism of Nb(V) deposition at BDD electrodes, a set of

potentiostatic experiments was performed fixing the potential at the

peak values determined in the cathodic scan of the cyclic

voltammetries, and it was imposed for 1 h under stirring conditions.

Figure 12 shows the SEM micrograph for the sample obtained fixing

the potential at -1.5 V along with the corresponding EDX analysis; a

black deposit was visible also with the naked eye; SEM pictures of

this sample reveal the presence of a layer on the BDD surface.


A deposit was not observed at -2.4 V and only traces of Nb were

detectable by EDX. At higher magnifications (Figure 13), a smooth

layer is well visible; the cracks can be attributed to the internal or

residual stress usually present in an electrodeposition process.

Figure 12: SEM micrograph and EDX profile of the electrodeposit formed

potentiostatically on BDD in [BMP][TFSA] containing 0.25 M NbF5 and

0.25 M LiF at potential of -1.5 V for 1 h at 125° C.

The EDX analyses show the presence of elemental Nb along with F,

O and C; in some cases also sulphur was detected. One assumption

is that sulphur and some of the oxygen may result from the ionic

liquid trapped in the crackle layer during the growth of the deposit.


Figure 14 shows the FIB cross section of a sample obtained under

potentiostatic conditions at -1.5 V for 10 minutes. An almost regular

coating was obtained, with a thickness of about 50 nm, which is in

agreement with the theoretical values calculated from the charge

passed in the potentiostatic experiment.

Figure 13: SEM micrograph of the electrodeposit formed potentiostatically

on BDD in [BMP][TFSA] containing 0.25 M NbF5 and 0.25 M LiF at

potential of -1.5 V for 1 h at 125° C.

Figure 14: SEM micrograph of the focused ion beam (FIB) cross section

of the electrodeposit formed potentiostatically on BDD in [BMP][TFSA]

containing 0.25 M NbF5 and 0.25 M LiF at potential of -1.5 V for 10

minutes at 125° C.


Element Atomic % concentration

C 61

O 12

F 6

Nb 21

Total 100

Table 1: Semi quantitative EDX elemental analysis of the Nb electrodeposit

obtained at -1.5 V for 3600 s.

The deposit obtained at -1.5 V was then characterized by XRD. The

specimens were prepared by imposing the potential for 15 h, in order

to obtain layer thick enough to be detected at wide angles.

A representative XRD pattern is shown in Figure 15 together with

the best-fitted Rietveld integral profile.

Figure 15: XRD pattern of layer deposited over the BDD substrate as a

function of the scattering angle 2θ. The Rietveld integral profiles belonging

to the different detected phases are shown together with the indexing

information (vertical bars below the XRD pattern).

The background line is also shown for reference. The experimental

data have been elaborated to enhance the visibility of XRD peaks at

wide angles. In particular, the scattered intensity I(Q) is multiplied


by the scattering wave vector Q, defined as 4π sin θ/λ, where is the

X-ray wave length. The logarithm of the product is plotted as a

function of the scattering angle 2. It can be seen that metallic Nb is

the dominant crystalline phase over the background mostly due to the

B-doped C diamond phase used as the conductive substrate during

the electrochemical deposition process. Secondary XRD reflections

can be ascribed to the formation of the stable carbide phase Nb2C,

and to the unusual one Nb6C5. It can be reliably concluded that almost

the 85% of deposited Nb is in its metallic form.

Taking these information into account it is possible to describe the

reduction of Nb(V) at BDD electrode from [BMP][TFSA]. Since

deposit was not obtained at -0.68 V, the first reduction peak observed

in the voltammetries can be attributed to the partial reduction of

Nb(V) to Nb(IV), Nb(III) or Nb(II). In pyrrolinidinium chloride ionic

liquid with addition of NbCl5, the formation of Nb(IV) and Nb(III)

chloride complex has been evidenced in the literature [40]. On the

other hand in melt salts the kind of anion determines the composition

of the first coordination sphere, stabilizing the higher oxidation state

when fluoride ions are involved: spectrochemical investigations

showed that in NaCl-KCl-K2NbF7 the Nb is mainly engaged in

fluorocomplexes such as NbF72- and NbF7

3- [45] [48] [49].

In the present work the electrochemical reduction has been made in

the presence of either NaF or LiF, so that fluorocomplex with Nb(IV)

may be formed as first step. Nevertheless, the presence of carbides as

Nb2C and Nb6C5 detected by X-ray diffraction are indicative of the

possible formation of Nb(II) and other non-stoichiometric forms. The

subsequent reduction peak centred at -1.3 V can be attributed to the

formation of Nb(0) since metallic Nb has been obtained by

potentiostatic deposition at -1.5 V. The anodic peak observed in

reverse scans at -1.1 V can be associated to the incomplete stripping

of the electrodeposit. However, the ratio of anodic to cathodic charge

is lower than one, revealing some irreversibility of this system.


Stripping seems to be kinetically hindered, which is a common

phenomenon in air and water stable ionic liquids [41]. Using LiF

instead of NaF the shift of the first and second reduction peaks can

be observed; this behaviour may be related to the role of cations of

the second coordination sphere, with a behaviour similar to that

observed in tantalum deposition from [BMP][TFSA]. The positive

effect of LiF on the quality of deposit was attributed to the ionic

polarizability of Li+ which can weaken the Ta–F bonds leading to a

facilitation of Ta deposition [50]. At the potential related to the third

reduction wave, deposit on the surface of BDD was not observed. A

similar behaviour was already observed during electrodeposition of

Ta from [BMP][TFSA] for which three reduction steps were

identified in the presence of LiF: the first and second steps were

related to the formation of metallic tantalum, while the third one was

connected to the formation of subhalides which can partially undergo

to further reduction to Ta [41] [51]. The formation of subhalides was

attributed to the accumulation of fluoride ions which were liberated

during the reduction steps: this accumulation increases the kinetics

of the formation of F-Ta complexes, favouring their formation with

respect to the reduction of Ta, although the latter process is

thermodynamically favoured, occurring at higher potential values.

Analogous phenomena may explain the electrochemical behaviour of

Nb observed in the present work: if the reduction of NbF5 is too fast,

as in the case of high applied overpotential, F- cannot diffuse fast

enough so that it accumulates near the surface, promoting the

formation of F-Nb complexes rather than the metallic Nb.

4.2.2. Polycrystalline gold substrate

Cyclic voltammograms of ionic liquid containing LiF and NbF5 on

polycrystalline gold at 125°C have been recorded at different scan

rates (Figure 16) and different potential windows (Figure 17).



NbF5 and 0.25 M LiF at polycrystalline gold electrode at 125° C .


NbF5 and 0.25 M LiF at polycrystalline gold electrode at 125° C (scan rate

100 mV/s).

-4

-3

-2

-1

0

1

2

3

4

-2.7 -1.7 -0.7 0.3 1.3

i /

mA

cm

-2


25 mV/s

50 mV/s

100 mV/s

200 mV/s

-3

-2

-1

0

1

2

-2.7 -1.7 -0.7 0.3 1.3

i /

mA

cm

-2



The electrode potential was scanned from the OCP in the negative

direction down to -2.5 V while the anodic scan was up to 1 V.

Two reduction waves can be observed in the cathodic scan at -1.2 V

and -1.6 V, while in the backward scan one oxidation peak and one

less defined oxidation wave appear at -0.2 V and 0.5 V, respectively.

A two-step mechanism can be proposed:

Nb(V) <=> Nb(III) <=> Nb(0)

Figure 18: SEM micrographs and EDX profile of the electrodeposit formed

potentiostatically on gold electrode in [BMP][TFSA] containing 0.25 M

NbF5 and 0.25 M LiF at potential of -1.6 V for 1 h at 125° C.


In order to get more information on the electrode process, the solution

was electrolyzed by holding the potential at -1.6 V resulting in a very

thin layer of a brown/black deposit on the working electrode.

The process at -1.2 V might be due to reduction of Nb(V) to Nb(III)

and to subvalent Nb species.

Agglomerated fine spherical structures with nanometric sizes are

seen in the SEM images (Figure 18); furthermore the EDX analysis

indicate the presence of some fluorine and oxygen in addition to Nb

in the deposit afterthorough washing: this may be due to trapped

electrolyte or due to the formation of subvalent NbFx films.

4.2.3. Copper substrate

The electrochemical behaviour of Nb was studied also using copper

substrates in [BMP][TFSA] containing 0.25 M NbF5 and 0.25 M LiF:

Figure 19 shows the voltammetric behaviour at different values of

scan rates.

The potential was scanned in the cathodic direction starting from the

OCP. Two reduction peaks are clearly evident at -0.5 V and -1.5 V,

indicating that the reduction of Nb(V) involves at least two

electrochemical steps under the conditions adopted in this work. In

the reverse scan two oxidation waves can be observed, although less

evident than the corresponding cathodic peaks. Also small waves at

-0.8 V and -1.8 V seem to appear at higher scan rate, but they cannot

be attributed to reactions involving Nb ions, as it can be observed in

the voltammograms obtained with the electrolyte prior the addition

of NbF5 (Figure 20), in which these peaks are evident.



NbF5 and 0.25 M LiF at copper electrode at 125° C.

Figure 20: Cyclic voltammogram of [BMP][TFSA] and 0.25 M LiF at

copper electrode at 125° C.

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

-2.5 -2 -1.5 -1 -0.5 0

i /

mA

cm-2

Potential / V vs. Pt-quasi ref.

25 mV/s

50 mV/s

100 mV/s

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

-2.5 -2 -1.5 -1 -0.5 0

i /

mA

cm-2


25 mV/s

50 mV/s

100 mV/s


Two waves appear at -0.5 V and -1.5 V confirming a two steps

reaction path Nb(V) => Nb(IV) => Nb(0); it is in agreement with the

results reported in the literature. To confirm this mechanism, a set of

potentiostatic experiments was performed; the potential was set at the

peak values determined in the cathodic scan of the voltammograms

and it was imposed for 1 h under stirring conditions.


potentiostatically on copper electrode in [BMP][TFSA] containing 0.25 M

NbF5 and 0.25 M LiF at potential of -1.5 V for 1 h at 125° C.


The copper sheets were then characterised: the SEM and EDX

analyses of the samples obtained at -0.5 V, which corresponds to the

potential of the first cathodic peak, reveal that elemental Nb is not

present on the surface.

Figure 21 shows the SEM micrographs for a sample obtained

potentiostatically at -1.5 V along with the corresponding EDX

analyses. A deposit is evident in the SEM pictures, while EDX

spectra reveal the presence of Nb. The SEM picture at higher

magnification of the deposit obtained at -1.5 V shows a particle

structure, with dimension in the range 50-100 nm.

Applying a potential more negative than -1.8 V, appreciable deposit

is not evidenced by SEM while EDX only detects traces of Nb.

Electrochemical deposition of tantalum


Figure 22 shows the cyclic voltammograms of [BMP][TFSA]

containing 0.25 M LiF and 0.25 M TaF5 recorded at BDD electrode

at different temperatures.

Scans were initially swept from the OCP towards the cathodic

direction at a rate of 100 mV s-1. The cyclic voltammograms exhibit

a different behaviour with increasing temperature: the current density

increases significantly in the range from 75 °C to 175 °C. This

behaviour may be related to the rise of the electrochemically active

species which leads to an increase of the redox reaction rates [41].

Two cathodic processes appear in the forward scan at -0.9 V and -1.8

V demonstrating that the electrochemical reduction of tantalum

apparently involves two steps, while in the reverse scan one oxidation

peak can be observed at 0.5 V. Extending the potential range, an

increase in the current densities at the cathodic and anodic limits is

observed; under these conditions the clear solution of [BMP][TFSA]

containing LiF changed to dark brown.


Figure 22: Cyclic voltammograms of 0.25 M TaF5 and 0.25 LiF in

[BMP][TFSA] ionic liquid at different temperatures (scan rate 100 mV/s).

The Ta(V) reduction has been studied in a wide range of melts, such

as chlorine or fluorine melts.

In FLINAK-K2TaF7 Polyakova et al. [32] have studied the nature of

the tantalum reduction process at 710 °C using various working

electrodes, such as molybdenum, platinum and silver; the authors

affirm that the cyclic voltammograms shape were not reproducible

and electrochemically TaF72- is reduced to tantalum metal in a single

quasi-reversible five-electron step.

In LiF-NaF-K2TaF7 with low oxygen contents on Ag electrode,

Chamelot et al. [33] recorded a single peak in the reduction sense

associated with one reoxidation peak on the reverse scan. The

reduction peak reveals the one-step reaction involving pentavalent

and zerovalen tantalum at 800 °C.

In chloride electrolytes Girginov et al. [28] have suggested that the

electrodeposition of microcrystalline Ta took place in a narrow

potential range and confirmed a single step mechanism in NaCl-KCl-

K2TaF7 melts. Zein El Abedin et al. [41] have shown the differences

-30

-25

-20

-15

-10

-5

0

5

-3 -2.3 -1.6 -0.9 -0.2 0.5 1.2

i /

mA

cm

-2


75 °C

125 °C

175 °C


on the electrochemical behaviour of [BMP][TFSA] ionic liquid with

different concentrations of TaF5 either in presence or absence of LiF

at platinum electrode; cyclic voltammograms clearly indicate that

three reduction peaks appear in the presence of LiF while only two

reduction peaks, correlated to the electrolytic reduction of Ta(V) to

Ta(III) and to the reduction of Ta(III) to Ta(0), have been recorded

without LiF addition. The two-step reduction mechanism of Ta(V)

was confirmed few years later by Borisenko et al. [21] that studied

the electrochemical behaviour of TaF5 in [Py1,4]TFSA on Au(111) at

25 °C. The electrochemical behaviour of tantalum(V) chloride and

oxochloride species has been also studied by Babushkina and Ekres

[42] in Pyr14Cl ionic liquid: the authors reported that the mechanism

of the electrochemical reduction of tantalum(V) depends strongly on

the composition of the electrolytic bath.


TaF5 and 0.25 M LiF at BDD electrode at 125° C.

Analysing the results obtained in the present work (Figure 23) the

first cathodic wave recorded at -0.9 V may be attributed to the

electrochemical reduction of Ta(V) to Ta(III), as TaF3 is known to be

-12

-10

-8

-6

-4

-2

0

2

4

-3 -2 -1 0 1 2

i /

mA

cm

-2


25 mV/s

50 mV/s

100 mV/s

200 mV/s


stable [21]. The second reduction peak at -1.8 V may be attributed to

the electroreduction of Ta(III) to Ta as at this electrode potential the

formation of a black deposit on the electrode surface is well visible

at naked eye. In the reverse scan two less-defined oxidation waves

can be observed; however, the ratio of anodic to cathodic charge is

lower than one, revealing that the reoxidation of the deposit is not

fully reversible.

Figure 24: Values of current density measured at the peak values of -1.1 V

and -1.8 V identified in the voltammograms obtained with BDD electrode

as a function of the square roots of the scan rates.

Plotting the current peaks as a function of the square roots of the scan

rates (Figure 24) a linear dependence is observed: this behaviour is

in agreement with the Randles-Sevčik equation (Equation 1), for a

planar diffusion controlled case [52], indicating that the reduction

process is mainly controlled by diffusion of the electroactive species

to the electrode surface.

0

2

4

6

8

10

0 5 10 15

i p/

mA

cm

-2

S.R.1/2 / mV1/2s-1/2

-1.8 V

-1.1 V


21

21

21

23

23

21

4463.0 Op cDTR

Fni Equation 1


potentiostatically on BDD in [BMP][TFSA] containing 0.25 M TaF5 and

0.25 M LiF at potential of -1.8 V for 1 h at 125° C.

Only when a potential of -1.8 V is applied, a dark-brown solid was

obtained which was characterised by SEM and EDX analyses (Figure

25). The results indicate a not completely uniform layer with

crystallites of size ranging from 50 to 100 nm. EDX profile reveals


the presence of Ta on the BDD surface; traces of F, O, C and S were

also detected probably originated from some trapped ionic liquid.

Element Atomic % concentration

C 9.59

O 10.04

F 28.93

Ta 51.44

Total 100

Table 2: Semi quantitative EDX elemental analysis of the Ta electrodeposit

obtained at -1.8 V for 3600 s.


Figure 26 shows the cyclic voltammograms of ionic liquid containing

LiF (0.25 M) and TaF5 (0.25 M) on polycrystalline gold at 125°C.


TaF5 and 0.25 M LiF at polycrystalline gold electrode at 125° C.

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

-3 -2 -1 0 1 2

i /

mA

cm

-2

Potential / V vs Pt quasi-ref.

25 mV/s

50 mV/s

100 mV/s

200 mV/s



potentiostatically on gold electrode in [BMP][TFSA] containing 0.25 M

TaF5 and 0.25 M LiF at potential of -1.8 V for 1 h at 125° C.

The electrode potential was scanned from the OCP at different scan

rates. The cyclic voltammograms exhibit two cathodic processes in

the forward scan at -0.5 V and -1.7 V, while in the backward scan

one oxidation peak appears. Analysing the results obtained in the

present work and taking into account the previous results showed for

BDD electrode, the first cathodic wave recorded at -0.5 V may be

attributed to electrochemical reduction of Ta(V) to Ta(III), while the


second reduction peak at -1.7 V may be attributed to the electrolytic

reduction of Ta(III) to Ta as at this electrode potential the formation

of a black deposit on the electrode surface is well visible at naked

eye. The anodic peak recorded on the backward scan is due to the

dissolution of the electrodeposit which, however, does not seem to

be complete.

The SEM micrograph of such a deposit made at 125°C at -1.8 V for

1 h is presented in Figure 27. As can be seen the electrodeposit

contains mainly particles and it is mainly constituted by Ta, but also

by C, O and F resulting from the remaining ionic liquid at the

electrode surface as revealed from the EDX profile.

4.3.3. Copper substrate

The behaviour of Ta(V) at copper electrode is presented in Figure 28:

two well-defined peaks at -0.8 V and -1.7 V have been recorded in

the ciclyc voltammograms. In the reverse scan a single oxidation

wave at -0.4 V is present. Also at copper electrode the peak current

densities (ip) related to the cathodic reactions of tantalum show a

linear dependence on the square root of the scan rates which confirms

the diffusion controlled nature of the process (Figure 29).

Also in this case, if the potential is kept at -1.8 V for 1 h, a black

deposit with a thin shining layer is obtained so that the second

reduction peak might be attributed to the reduction from Ta(III) to

Ta. The anodic peak that appears at -0.4 V may be attributed to the

partial stripping of the electrodeposit. However, the ratio of anodic

to cathodic current is much lower than 1, indicating some

irreversibility of the system.



TaF5 and 0.25 M LiF at copper electrode at 125° C.

Figure 29: Values of current density measured at the peak values of -0.8 V

and -1.7 V identified in the voltammograms obtained with copper electrode

as a function of the square roots of the scan rates.

-15

-12

-9

-6

-3

0

3

6

-2.5 -2 -1.5 -1 -0.5 0

i /

mA

cm

-2


25 mV/s

50 mV/s

100 mV/s

200 mV/s

0

2

4

6

8

10

12

14

16

0 5 10 15

i p/

mA

cm

-2

S.R.1/2 / mV1/2 s-1/2

-1.7 V

-0.8 V



potentiostatically on copper electrode in [BMP][TFSA] containing 0.25 M

TaF5 and 0.25 M LiF at potential of -1.8 V for 1 h at 125° C.

Figure 30 shows the SEM and EDX results of a deposit obtained on

copper surface fixing the potential at -1.8 V for 1 h at 125 °C under

stirring conditions. EDX analysis shows the presence of fluorine,

oxygen and carbon besides tantalum, resulting from the ionic liquid

trapped in the layer during the growth of the deposit; very rarely

sulphur was detected. As can be seen, the layer is quite rough and not

completely uniform; high resolution micrograph shows the growth of

some particles in the range of some ten nanometers.


In order to determine the possible reaction path for tantalum

electrodeposition a set of chronopotentiometry experiments have

been performed at copper electrode at 125 °C under different applied

current intensities (Figure 31).

Figure 31: Trend with time of potential measured during galvanostatic

experiments at different values of imposed current density. Inset shows a

magnification of the first 2 seconds of galvanostatic experiments.

Two plateaux can be observed at -1.2 V and -1.8 V confirming that

the electroreduction of Ta(V) is a diffusion controlled process in two-

steps. The number of electrons exchanged in each step can be

determined by analysing the transition times (1 and 2) that

correspond to the observed potential plateaux. By using the Sand

equation [53] for a two steps in-series electrochemical reaction, it is

possible to correlate the first transition time (1) with the imposed

current, the diffusivity and the concentration of the starting ion and

the number of exchanged electrons (n1). The second transition time

(2) depends on the same parameters of the first, but it is correlated

with the number of electrons exchanged in the second step (n2), and

-2.1

-1.9

-1.7

-1.5

-1.3

-1.1

-0.9

-0.7

0 5 10 15

Po

ten

tia

l /

V

Time / s

-3.25 mA -3.5 mA -3.75 mA

-2.1

-1.9

-1.7

-1.5

-1.3

-1.1

-0.9

-0.7

0 0.5 1 1.5 2

Po

ten

tial

/ V

Time / s


with diffusivity and concentration of the electroactive species formed

in the first step. The specific case of two steps in-series reactions

diffusion controlled has been studied by Berzins and Delahay [53],

which provided the following equation which correlates 1 and 2:

𝜏2 = 𝜏1 [2 (𝑛2

𝑛1) + (

𝑛2

𝑛1)

2

] Equation 2

Table 3 reports the values of transition times obtained at copper

electrode under different applied current densities. The average 2/1

approaches 5.25 which is in agreement with the theoretical value

obtainable with Equation 2 for n1=2 and n2=3. Taking these

information into account, a possible reaction path for tantalum

reduction is:

Ta(V) => Ta (III) => Ta

i (mA cm-2) 1 (s) 2 (s) 2/1

-5.41 1.28 8.50 6.64

-5.83 1.00 5.28 5.28

-6.25 0.78 3.52 4.52

Table 3: Values of transition times under different applied current densities

Electrochemical deposition of zirconium


Cyclic voltammetries have been performed at 125°C in 0.25 M LiF

and 0.25 M ZrF4 [BMP][TFSA] (Figure 32); the potential was

scanned in the cathodic direction, starting from the OCP.



ZrF4 and 0.25 M LiF at BDD electrode at 125° C.


ZrF4 and 0.25 M LiF at BDD electrode at 125° C (scan rate 100 mV/s).

-4

-3

-2

-1

0

1

2

-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5

i /

mA

cm

-2


25 mV/s50 mV/s100 mV/s200 mV/s

-4

-3

-2

-1

0

1

-3 -2 -1 0 1 2

i /

mA

cm

-2



Figure 33 shows the voltammograms at 100 mV s-1 with different

potential ranges. As can be seen, two reduction waves appear at -1.3

V and -1.85 V and two oxidation waves can be observed at -1.3 V

and 0.5 V. Under the operative conditions adopted in this work, the

reduction of Zr(IV) in [BMP][TFSA] seems to occur in a two-step

mechanism. However, the attribution of peaks is not straightforward.

The electrochemical behaviour of ionic liquids at different materials

has been already observed by other authors, but not completely

understood [21].

The effect of fluoride ions on the reduction of zirconium has been

studied by Guang-Sen et al. [34] in fluoride-chloride melts: the

voltammetric behaviour of zirconium at 973 K and 1033 K in

equimolar KCl-NaCl with ZrCl4 on a platinum electrode showed that

the electroreduction process of Zr is reversible. According to the

number of electrons exchanged, the authors established that the

process has double two-electron steps: Zr(IV)/Zr(II) and Zr(II)/Zr.

The cathodic reduction of Zr(IV), in the same electrolytic bath, was

also studied with different concentrations of KF: a new peak

appeared when small amount of KF (XF-/XZr

4+ = 1.07) was added into

equimolar KCl-NaCl melts with ZrCl4. The currents at peaks related

to Zr(IV)/Zr(II) and Zr(II)/Zr reduction decreased as the ratio (XF-

/XZr4+) in the melts increased. The authors have concluded that the

new peak recorded in the presence of KF represents the main

reduction of Zr when the ratio XF-/XZr

4+≥ 6.

Kawase and Ito [35] have studied the voltammetric behaviour at Ni

electrode in the solution obtained after the anodic dissolution of Zr in

the LiCl-KCl eutectic. Cyclic voltammograms reveal that Zr(II) and

Zr(IV) can exist in molten salts between 450 and 550 °C and the ratio

of Zr(II)/Zr(IV) increases with increasing temperature. At 450 °C

only one cathodic peak related to Zr(IV)/Zr(II) is observed, while at

550 °C two cathodic peaks, related to the reduction of Zr(IV)/Zr(II)

and Zr(II)/Zr, are observed.


The electroreduction process of Zr(IV) was studied by Chen et al.

[36] at molybdenum electrode in LiCl-KCl-K2ZrF6 molten salts at

923 K; the authors highlighted that Zr(IV) was reduced to Zr metal

by a two-step mechanism corresponding to the Zr(IV)/Zr(II) and

Zr(II)/Zr transitions. Furthermore, the cathodic peaks are not

dependent on the sweep rate so it has been concluded that the

reduction processes of Zr(II)/Zr and Zr(IV)/Zr(II) are both reversible

(or quasi-reversible).

Analogous behaviour has been reported by Wu et al. [37] that studied

the electrochemical reduction of Zr(IV) on Pt electrode at 1023 K in

NaCl-KCl-K2ZrF6 molten salt; two cathodic reduction peaks related

to Zr(IV)/Zr(II) and Zr(II)/Zr steps were observed in the potential

range between 0 and -1.6 V vs. Ag/AgCl. The authors thus observed

a two-electron transfer process involving the formation of Zr(II).

On the other hand, Groult et al. [38] studied the electrochemical

behaviour of Zr(IV) in potassium-free fused alkali fluorides in the

temperature range between 773 and 1123 K by using either tungsten

or molybdenum electrodes, and observed a single well-defined

oxido-reduction wave, and concluded that the reduction of Zr(IV) is

a single four-electron reversible step controlled by diffusion process.

A detailed electrochemical study of the molten LiF-CaF2-ZrF4

system is provided in the 810-920 °C temperature range by Gibilaro

et al. [39] on an inert Ta electrode. The authors have reported that at

840 °C a single peak is observed in the cathodic run associated with

a stripping peak; the reduction mechanism was a one step process

exchanging 4 electrons and controlled by the Zr ions diffusion in the

molten salt.

Moreover, the electrolyte may originate voltammetric responses:

Randstrom et al. attributed to traces of oxygen and water in the

nitrogen atmosphere the presence of peaks in the voltammograms

[54] [55].


Figure 34: SEM micrographs and EDX profile of the deposit obtained at

-1.85 V for 1 h at 125° C on BDD in [BMP][TFSA] containing 0.25 M ZrF4

and 0.25 M LiF.

In order to study the possible reduction mechanism for Zr(IV),

different potentiostatic experiments were performed setting the

potential at the peak values determined in the cathodic scan of the

cyclic voltammetries, for 1 h under stirring conditions.

The samples obtained were characterized by SEM-EDX. SEM and

EDX analyses revealed that Zr is not present on the surface of


samples obtained at potential corresponding to the first cathodic

peak. At potential values corresponding to -1.85 V, the SEM pattern

(Figure 34) shows the presence of a solid deposit: EDX analyses

reveal the presence of Zr along with F, which may derive from ionic

liquid trapped in the deposit.

In order to determine if the deposit obtained on the BDD surface is

constituted by metallic zirconium, X-ray diffraction (XRD) have

been performed: Figure 35 shows an example of XRD spectra. The

XRD pattern in the angular range 25-75°, shows diffraction peaks

marked with black stars, that correspond to the hexagonal structure

of zirconium with a space group P63/mmc (ICSD# 05-0665).

Figure 35: XRD pattern of the deposit obtained at -1.85 V for 1 h at 125

°C on BDD as a function of the scattering angle 2.

The slight shift with respect to the value reported in the ICSD may

derive from lattice distortion, originated by the fluorides arising from

unreacted ZrF4 [56].

The deposit obtained at -1.85 V is then constituted by metallic Zr,

and as showed by SEM analyses, is characterised by fine crystallites

with average sizes in the range 50-100 nm.


Under the conditions adopted in this work, the reduction of Zr

involves two electrochemical steps:

Zr(IV) => Zr(II) => Zr


The electrochemical behaviour of Zr(IV) was investigated also at

polycrystalline gold as substrate.

When ionic liquid contains only 0.25 M of LiF small peaks were

present, with current densities lower than 0.5 mA cm-2 (Inset Figure

36). In the solution containing both LiF and ZrF4 two well defined

cathodic peaks at -1.0 V and -1.4 V appear at noble metal electrode;

in the backward scan only one defined oxidation peak at -0.38 V and

a shoulder are evident (Figure 36).

The voltammetric study clearly shows two electrochemical steps

involved in the Zr(IV) reduction: the attribution of the peaks has been

done on the basis of chronopotentiometries and surface analyses,

which are discussed below.

As can be seen from Figure 37, if the reduction scan is stopped at

-1.4 V the oxidation peak at -0.38 V is still present while it does not

appear in the voltammogram stopping the cathodic scan at -0.8 V.

A two-steps mechanism was also observed in molten salts in which

Zr(IV), Zr(II) and Zr(0) have been found [57] [58] as intermediates

and final products. However, depending on temperature and

composition of the bath, also the direct reduction to Zr(0), involving

four electrons was found as possible reaction mechanism [34] [38]

[39]. In pyrrolinidinium based ionic liquid at room temperature, also

Zr(III) has been individuated during the reduction process [44].



ZrF4 and 0.25 M LiF at polycrystalline gold electrode at 125° C (Inset

shows cyclic voltammogram of [BMP][TFSA] containing 0.25 M LiF at

125 °C, scan rate 100 mV/s ).


ZrF4 and 0.25 M LiF on polycrystalline gold electrode at 125° C (scan rate

100 mV/s).

-4.0

-3.0

-2.0

-1.0

0.0

1.0

2.0

-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5

i /

mA

cm

-2


25 mV/s

50 mV/s

100 mV/s

200 mV/s

-3.0

-1.5

0.0

1.5

3.0

-3.2 -2.2 -1.2 -0.2 0.8 1.8

i /

mA

cm

-2Potential / V vs. Pt-quasi ref.

-3.0

-2.0

-1.0

0.0

1.0

-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5

i /

mA

cm

-2



Figure 38: SEM micrographs and EDX profiles of the deposit obtained at

-1.85 V for 1 h at 125 °C on gold electrode in [BMP][TFSA] containing

0.25 M ZrF4 and 0.25 M LiF.

Moreover, a detailed study conducted on monocrystalline (111) gold

electrode in pyrrolidinium based ionic liquid showed some

characteristic peaks related to the strong interaction between the

pyrrolinidium cations and the electrode surface [59].


Potentiostatic experiments were then performed to characterise the

final reduction products, setting the potential at the peak values,

determined in the cathodic scan of the cyclic voltammetries, for 1 h

under stirring conditions. The samples obtained were characterized

by SEM-EDX. SEM and EDX analyses revealed that Zr is not present

on the surface of samples obtained at potential corresponding to the

first cathodic peak. At potential values corresponding to -1.45 V, the

SEM pattern shows the presence of a solid deposit: EDX analyses

reveal the presence of Zr along with F, which may derive from ionic

liquid trapped in the deposit (Figure 38).

Figure 39: Trend with time of potential measured during galvanostatic

experiments at different values of imposed current density at gold electrode

at 125 °C.

To sketch a possible reaction path for the Zr deposition, a set of

chronopotentiometries and potentiostatic experiments have been

performed in this work. Figure 39 shows an example of

-3

-2.5

-2

-1.5

-1

-0.5

0

0 5 10 15 20

Pote

nti

al

/ V

vs.

Pt-

qu

asi

ref

.

time / s

2.0 mA 1.7 mA 1.5 mA

-2

-1.5

-1

-0.5

0

0 1 2 3

Pot

enti

al /

V v

s. P

t-qu

asi

ref.

time / s


cronopotentiogram obtained at gold electrode: the experiments were

carried out at different current intensity at 125°C.

Two plateaux can be observed at -1.0 V and -1.45 V corresponding

to the peak potentials of the cyclic voltammograms at gold electrode,

confirming that the electroreduction of Zr(IV) at gold electrode

proceeds via two-step processes, and allow to establish that these

steps are diffusion-controlled.

From the analysis of the transition times (1 and 2) corresponding to

the observed potential plateaux, it is possible to obtain information

about the number of electrons exchanged in each step (Equation 2).

For a two steps in-series electrochemical reaction, the first transition

time (1) can be correlated with the imposed current, the diffusivity

and the concentration of the starting ion and the number of exchanged

electrons (n1) by the Sand equation [53]. The second transition time

(2) depends on the same parameters of the first; moreover it is

correlated with the number of electrons exchanged in the second step

(n2), and with diffusivity and concentration of the electroactive

species formed in the first step.

i (mA cm-2) 1 (s) 2 (s) 2/1

-2.50 3.2 10.4 3.25

-2.83 2.2 6.9 3.13

-3.33 1.45 4.13 2.84

Table 4: Values of transition times at different electrode under different

applied current densities.

Table 4 reports the values of transition times obtained for each

electrode under different applied current densities. The average 2/1

value approaches 3, which is in agreement with the theoretical value

obtainable with Equation 2 for n1 = n2. Taking into account the

possible valence numbers of zirconium, the electrons exchanged in

the two-steps reduction could be 1 or 2. Since metallic Zr was


obtained as final product of the reduction process, the results allow

to establish the following reaction path:

Zr(IV) => Zr(II) => Zr

Electrochemical deposition of copper

4.5.1. Dissolution of pure copper metal

The electrochemical behaviour of copper ions was studied in the

solution of ionic liquid prepared introducing copper cations into

[BMP][TFSA] by anodic dissolution of pure copper metal.

The study of the electrochemical dissolution of copper electrode is

presented in Figure 40: the steady-state curve, measured under

stirring conditions at 125 °C, shows a single wave at ca. 0.1 V

indicating that the dissolution of copper occurs through a single

electrochemical step under the conditions adopted in this work.

In order to determine the oxidation state of copper species in

[BMP][TFSA], the weight loss of the copper electrode and the charge

passed during its anodic dissolution at a potential of 0.1 V were

measured: a value of 49.14 C corresponding to a change in weight of

33 mg allows to confirm that the oxidation state of copper species is

1. This copper form may presumably be stabilized by the TFSA anion

as CuTFSA. As reported in different works, Cu(I) species present

high stability in ionic liquids such as 1-butyl-1-methylpyrrolidinium

bis(trifluoromethylsulfonyl)amide [60] 1-ethyl-3-

methylimidazolium tetrafluoroborate [61], and trimethyl-n-

hexylammonium bis(trifluoromethylsulfonyl) amide [62]; the

anodization of metallic copper in these ionic liquids always led to

dissolved monovalent Cu.


Figure 40: Steady-state response recorded in [BMP][TFSA] at copper

electrode at 125° C (scan rate 0.5 mV/s).

4.5.2. Niobium substrate

Figure 41 shows the cyclic voltammograms obtained for Cu(I)

species at niobium electrode at different scan rates in [BMP][TFSA]

without adding LiF: the potential was scanned in the cathodic

direction from the OCP to -1.4 V. As can be seen, all cyclic

voltammograms exhibit the same general features with one cathodic

peak and its corresponding anodic peak.

This behaviour is in agreement with the results obtained for the

dissolution of copper: in the cathodic scan the reduction of Cu(I) to

metallic copper occurs while in the reverse scan the stripping peak of

the metal is visible. Moreover, when the potential scan is reversed

after peak recorded at -0.8 V, a current loop is visible which is

indicative of the nucleation process of copper on niobium surface

[61].

0.0

2.0

4.0

6.0

8.0

10.0

12.0

-0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4

i /

mA

cm

-2

Potential /V vs. Pt-quasi ref.


As the effect of scan rate is concerned, it can be observed that, while

the anodic peak is centred at -0.06 V, the cathodic peak shifts

negatively with the scan rate from -0.62 to -0.87 V. In the examined

range of scan rates the system shows then a quasi-reversible

behaviour with a rate of electron transfer for copper deposition

comparable to the rate of mass transfer.

Figure 41: Cyclic voltammograms of [BMP][TFSA] containing dissolved

Cu at niobium electrode at 125 °C.

Figure 42 reports the cyclic voltammogram obtained for Cu(I)

species at niobium electrode: in this case the potential was scanned

in the cathodic direction from the OCP to -1.4 V and the anodic stop

potential has been increased up to 0.5 V.

After the stripping peak at -0.1 V the current density increases, and

in the backward scan a new peak is present at ca. -0.3 V, indicating

that Cu(II) species have been formed from the oxidation of Cu(I),

which can be reduced by two one-electron steps in the cathodic

sweep. Also in this case the current loop typical of a nucleation

process is visible.

-15.0

-10.0

-5.0

0.0

5.0

10.0

15.0

20.0

-1.5 -1 -0.5 0 0.5

i /

mA

cm

-2


25 mV/s

50 mV/s

100 mV/s

200 mV/s


Figure 42: Cyclic voltammogram of [BMP][TFSA] containing dissolved

Cu at niobium electrode at 125° C (scan rate 100 mV/s).

Experiments were also performed using [BMP][TFSA] with 0.25 M

LiF in order to study the behaviour of copper in a solvent similar to

that required for the refractory metals deposition.

As can be seen from Figure 43, the voltammetric behaviour

significantly differs when LiF is added to the ionic liquid during the

dissolution of copper: scanning the potential in the cathodic direction

from the OCP to -1.4 V, two cathodic peaks are present and two not

well-resolved anodic peaks are visible in the reverse scan. Moreover,

the peak potentials are shifted to values more negative than the

corresponding values obtained without LiF.

The presence of two reduction peaks in the first cathodic scan

indicates that in the dissolution process both Cu(I) and Cu(II) are

formed. However, the steady-state curves of dissolution of copper

electrode along with the measurements of charge passed in the

anodization vs. weight loss clearly indicate a one-electron process to

form Cu(I). The presence of Cu(II) species in this solution can be

-20.0

-15.0

-10.0

-5.0

0.0

5.0

10.0

15.0

20.0

25.0

30.0

-1.5 -1 -0.5 0 0.5 1

i /

mA

cm

-2


100 mV/s


thus ascribed to the presence of fluoride ions in the solution: Cu(I)

electrochemically produced could be stabilized by TFSA complex,

but in presence of fluorides CuF2 may be formed, allowing to a

solution containing the two copper species [63].

The first cathodic peak in the voltammograms can be then attributed

to the reduction of Cu(II) fluoride to Cu(I) at -0.5 V followed by the

reduction wave of Cu(I) at -1.0 V, while a stripping peak in the

reverse scan pertains to the electrochemical formation of Cu(I).


Cu and 0.25 M LiF at niobium electrode at 125° C.

The deposition of copper on niobium electrode was achieved in the

ionic liquid [BMP][TFSA] containing Cu(I) and 0.25 M LiF at 125

◦C , at a fixed potential of -0.8 V for 1 h. Under these conditions, an

adhesive red-brown deposit with metallic luster was visible on the

electrode surface.

-4.0

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0

4.0

-1.5 -1 -0.5 0 0.5

i /

mA

cm

-2


25 mV/s

50 mV/s

100 mV/s

200 mV/s


The SEM micrograph and EDX analyses in Figure 44 show the

surface morphology of an electrodeposited copper layer on niobium

substrate: the deposit is uniform and contains fine crystallites with

average sizes of about 100 nm. EDX analyses in different points of

the deposit indicate that Cu covers the whole Nb surface.


potentiostatically on niobium substrate from [BMP][TFSA] containing Cu

and 0.25 M LiF at potential of -0.8 V for 1 h at 125° C.


4.5.3. Tantalum substrate

Similar results were obtained using tantalum foils instead of niobium

foils as working electrode.

Figure 45 reports the electrochemical behaviour of copper at

tantalum substrate: one reduction peak at -0.8 V and the current loop

typical of a nucleation process are visible in the voltammograms.

The potentiostatic deposition has been performed at a fixed potential

of -0.8 V for 1 h: a red-brown deposit was visible on tantalum

substrate also with naked eyes.

SEM micrograph and EDX analyses (Figure 46) reveal that the

deposit is uniform and contains fine crystallites with average size

ranging from 50 to 100 nm; EDX analyses indicate that copper covers

almost completely the Ta surface.


Cu and 0.25 M LiF at tantalum electrode at 125 °C.

-2.0

-1.5

-1.0

-0.5

0.0

0.5

-1.5 -1 -0.5 0 0.5

i /

mA

cm

-2


25 mV/s

50 mV/s

100 mV/s

200 mV/s



potentiostatically on tantalum electrode in [BMP][TFSA] containing

dissolved Cu and 0.25 M LiF at potential of -0.8 V for 1 h at 125° C.

Electrochemical deposition of composites

The mechanical properties of metallic multilayer composites have

drawn world-wide attention in recent years. More specifically, fine-

scale composites such as Cu/Nb and Cu/Ta have been the subject of

many recent experimental and theoretical studies [64]. These fine


scale composite materials typically exhibit high yield strengths,

which at room temperature can approach 1/2 to 1/3 of the theoretical

strength.

Cu/Nb and Cu/Ta multilayer composites are immiscible systems and

have been reported to exhibit length scale effect in some properties

such as electrical resistivity and fracture properties [65].

4.6.1. Electrodepostion of Cu/Nb composites

Cu/Nb composites were prepared in dual bath using niobium foils as

working electrode. A first deposit of copper was prepared at -0.8 V

for 1800 s, followed by a niobium deposit obtained at -1.5 V for 3600

s. This procedure was repeated twice and the resulting sample was

characterized by SEM-EDX analyses.

The results are reported in Figure 47 and Figure 48: the deposit is

regular and the EDX analysis shows the presence of both copper and

niobium. As it can be observed form the elemental maps, the last

deposit of niobium is not uniform and copper is still visible.

Element Wt%

C 5.46

O 13.92

F 4.13

Cu 44.15

Nb 32.33

Total 100

Table 5: Semi quantitative EDX elemental analysis of the Cu/Nb

composites obtained in dual bath mode on niobium electrode: first deposit

of Cu prepared at -0.8 V for 1800 s, followed by Nb deposit obtained at

-1.5 V for 3600 s. The procedure was repeated twice.


Figure 47: SEM micrograph and elemental maps of the Cu/Nb

electrodeposit obtained in dual bath mode on niobium electrode: first

deposit of Cu prepared at -0.8 V for 1800 s, followed by Nb deposit obtained

at -1.5 V for 3600 s. The procedure was repeated twice.

Figure 48: EDX profile of the Cu/Nb electrodeposit obtained in dual bath

mode on niobium electrode: first deposit of Cu prepared at -0.8 V for 1800

s, followed by Nb deposit obtained at -1.5 V for 3600 s. The procedure was

repeated twice.

Cu

Nb Cu/Nb


The semiquantitative elemental analysis of the sample (Table 5)

reports comparable amounts of niobium and copper: the presence of

14% of oxygen can be related to copper and niobium oxide formed

at the surface of the sample exposed to air.

The FIB micrograph (Figure 49) of this sample shows that the first

layer of copper is regular and dense, while the subsequent deposition

of niobium presents a more porous structure. The further deposition

of copper occurs on this porous surface, as well as the last niobium

deposit.

Figure 49: SEM micrographs of the focused ion beam (FIB) cross section

of the Cu/Nb electrodeposit obtained in dual bath mode on niobium

electrode: first deposit of Cu prepared at -0.8 V for 1800 s, followed by Nb

deposit obtained at -1.5 V for 3600 s. The procedure was repeated twice.

4.6.2. Electrodepostion of Cu/Ta composites

The electrodeposition of Cu/Ta composites has been carried out in

dual bath using either BDD substrate (Figure 50 and Figure 51) or

tantalum foils (Figure 52) as working electrode. The first layer of

copper was realized fixing the potential at -0.8 V for 1800 s, followed

by a tantalum deposit obtained at -1.8 V for 3600 s. Before the

electrodeposition of tantalum, the sample, coated with the copper

layer, was washed in isopropanol in order to avoid the contamination


of the solution of IL containing LiF and TaF5. The procedure was

repeated twice.

Figure 50: SEM micrograph and EDX profile of the Cu/Ta electrodeposit

obtained in dual bath mode on BDD electrode: first deposit of Cu prepared

at -0.8 V for 1800 s, followed by Ta deposit obtained at -1.8 V for 3600 s.

Figure 51: SEM micrograph and elemental maps of the Cu/Ta

electrodeposit obtained in dual bath mode on BDD electrode: first deposit

of Cu prepared at -0.8 V for 1800 s, followed by Ta deposit obtained at

-1.8 V for 3600 s.

Cu

Ta Cu/Ta



electrodeposit obtained in dual bath mode on tantalum electrode: first

deposit of Cu prepared at -0.8 V for 1800 s, followed by Ta deposit obtained

at -1.8 V for 3600 s.

The FIB micrograph of the composite realized on tantalum substrate

(Figure 53) shows that the first layer of copper is almost regular with

average size of 50 nm, while the subsequent deposition of tantalum

presents a more porous structure. The further deposition of copper

and tantalum occurs on this porous surface.

Ta Cu/Ta

Cu

7


Figure 53: SEM micrograph and EDX analysis of the focused ion beam

(FIB) cross section of the Cu/Ta electrodeposit obtained in dual bath mode

on tantalum electrode: first deposit of copper prepared at -0.8 V for 1800

s, followed by tantalum deposit obtained at -1.8 V for 3600 s.

Conclusions

In this Chapter conclusions are given.

92 Chapter 5. Conclusions

Based on the objectives introduced at the beginning of this thesis,

several conclusions can be drawn from this study.

The pre-electrolysis process implemented in this study was shown to

be effective in removing impurities from the as-received ionic liquid.

The electroreduction of NbF5, TaF5, ZrF4 was investigated in the

purified ionic liquid, 1-butyl-1-methyl-pyrrolidinium bis(tri-

fluoromethylsulfonyl)imide at 125 °C.

The copper solutions were obtained by dissolution of metallic

copper: cyclic and linear sweep voltammetries showed that both

Cu(I) and Cu(II) may be present, depending on the electrolyte

composition.

Under potentionstatic conditions thin coatings of Nb, Ta, Zr and Cu

were obtained in the presence of LiF; however further

electrochemical experiments are required to improve the

electrochemical parameters in order for a uniform, adherent layer of

metals to be obtained.

The samples were characterized by SEM-EDX and XRD analyses.

SEM images reveal that a deposit of niobium can be obtained at

-1.5 V, while at -0.7 V and -2.4 V niobium is not present on the BDD

surface; the XRD pattern shows that metallic Nb is the dominant

crystalline phase over the background mostly due to the B-doped C

diamond phase used as the conductive substrate during the

electrochemical deposition process. Also using a metal substrate, a

deposit of niobium was observed at -1.5 V at 125°C.

On the electrodeposition of tantalum is concerned, only when a

potential of -1.8 V is applied a dark-brown solid was obtained on

BDD, gold and copper substrates.

Metallic zirconium was obtained fixing the potential at the value

corresponding to the more negative reduction peak that appears in the

voltammograms recorded using different substrates.

Chapter 5. Conclusions 93

The EDX analyses show the presence of the metal of interest along

with F, O and C; in some cases also sulphur was detected. One

assumption is that sulphur and some of the oxygen may result from

the ionic liquid trapped in the crackle layer during the growth of the

deposit.

Copper-niobium and copper-tantalum nanometric composites were

prepared in dual bath using different working electrodes.

The deposits were characterised by SEM-EDX analyses revealing

that both Cu/Nb and Cu/Ta composites are characterised by fine

crystallites with average sizes in the range 50-100 nm. Although the

results are promising, further studies are requested to improve the

morphological properties of the resulting composites.

The present work has shown that it is possible to deposit a layer of

niobium, tantalum, zirconium and copper from an ionic liquid which

underwent a specialised pre-electrolysis process. However, further

work is recommended in improving the adherence and uniformity of

the layers on the substrates.

Appendix

Electrochemical background

and techniques

This chapter is focused on the electrochemical processes taking place

at electrode surface and electrochemical techniques used to

investigate the electrochemical behaviour of electrode materials.

96 Appendix. Electrochemical background and techniques

Electrochemical processes

Electrochemical reactions are heterogeneous chemical processes

which involve the transfer of charge in a ionic conductor (electrolyte)

to or from a surface (electrode), generally a metal or a semiconductor.

Charge is transported through the electrode by the movement of

electrons. In the electrolyte phase, charge is provided by the

movement of ions. Electrochemical cells are defined most generally

as two electrodes (anode and cathode) separated by at least one

electrolyte phase.

The electrochemical reaction taking place in a cell is the result of two

independent half-reactions, which describe the real chemical changes

at the two electrodes. The reaction that take place at the electrode of

interest (working electrode) can be studied by using a second

electrode (counter electrode). The potential of the working electrode

can be controlled with respect to the reference electrode; since

reference electrode has a fixed potential it is possible to relate any

changes in the cell to the working electrode.

Figure 54: Schematic diagram of an electrochemical cell.

Appendix. Electrochemical background and techniques 97

In a typical electrochemical experiment were a working and a

reference electrode are immersed in a solution, the variation of the

potential, E, by means of an external power supply, can produce a

current flow in the external circuit because, as reactions occur,

electrons cross the electrode/solution interfaces.

Moreover, when electrolysis occurs, in addition to electron transfer

at the anode and cathode surfaces, ions must pass through the solution

between the electrodes and electrons though the wires externally

interconnecting the two electrodes (in order to maintain electrical

neutrality at all points in the system).

Hence the current through the external circuit, I, given by

𝐼 = 𝐴 𝑖

(when A is the electrode area and i the current density), is a

convenient measure of the rate of the cell reaction.

The number of electrons that cross an interface is measured in terms

of the total charge, Q, passed in the circuit. Moreover, the current

through the external circuit is a measure of the rate of the cell

reaction.

The relationship between charge and amount of product formed is

given by Faraday's law.

𝑄 = ∫ 𝐼 𝑑𝑡𝑡

0

= 𝑚𝑛𝐹

where Q is the charge, passed during a period t, required to convert

m moles of starting material to product in an electrode reaction

involving the transfer of n electrons/molecule; F is 96485.4 С.

Two types of processes occur at electrodes: one kind involves

processes in which charges are transferred across the electrode-

solution interface. Since such reactions are governed by Faraday's

law they are called faradaic processes. Electrodes at which faradaic


processes occur are sometimes called charge transfer electrodes.

Under some conditions, a given electrode-solution interface will

show a range of potentials where no charge-transfer reactions occur

because such reactions are thermodynamically or kinetically

unfavourable. However, processes such as adsorption and desorption

can occur, and the structure of the electrode-solution interface can

change with changing potential or solution composition. These

processes are called nonfaradaic processes. Although charge does not

cross the interface, external currents can flow (at least transiently)

when the potential, electrode area, or solution composition change.

Both faradaic and nonfaradaic processes occur when electrode

reactions take place. Although the faradaic processes are usually of

primary interest in the investigation of an electrode reaction, the

effects of the nonfaradaic processes must be taken into account in

using electrochemical data to obtain information about the charge

transfer and associated reactions.

A.1.1. Nonfaradaic processes

An electrode at which no charge transfer can occur across the metal-

solution interface, is called an ideal polarized electrode (IPE). While

no real electrode can behave as an IPE over the whole potential range

available in a solution, some electrode-solution systems can approach

ideal polarizability over limited potential ranges [53].

At any interface between two phases and particularly between an

electrode and an electrolyte solution, there exists a segregation of

positive and negative charge in a direction normal to the phase

boundary. These charges may be associated in the form of dipolar

molecules or polarised atoms, or they may be free as holes, electrons,

or ions. The charge segregation may occur through the preferential

adsorption of either positive or negative ions at the interface, through

the transfer of charge across the interface, or through the deformation

of polarisable molecules by the asymmetrical force field at the

interface. The theory of the electrical double layer is concerned with


the charge distribution and electrical potentials that arise as a

consequence of this charge separation [52].

The study of the electrical double layer is intimately concerned with

the concept of the ideally polarisable electrode; since, at ideally

polarisable electrode, charge does not cross the interface during

potential changes, the interface behaves like a capacitor. For a given

applied potential, charge is stored in equal measure on the electrode

and in the solution. In the case of a metallic electrode, the excess or

deficiency of electrons is limited to a very thin surface layer, while

the charge in solution is distributed in the proximity of the electrode

surface. This charge distribution at the interface is called double layer

with which a double-layer capacitance (Cdl) is associated.

Figure 55: Proposed model of the double-layer region under conditions

where anions are specifically adsorbed.

The solution side of the double layer is constituted of several

"layers". That closest to the electrode, the inner layer, called also the

Helmholtz layer, contains solvent molecules and sometimes other


species (ions or molecules). The locus of the electrical centres of the

specifically adsorbed ions is called the inner Helmholtz plane (IHP),

which is at a distance x1. Solvated ions can approach the metal only

to a distance x2; the locus of centres of these nearest solvated ions is

called the outer Helmholtz plane (OHP). The interaction of the

solvated ions with the charged metal involves only long-range

electrostatic forces, so that their interaction is essentially independent

of the chemical properties of the ions. These ions are said to be

nonspecifically adsorbed. Because of thermal agitation in the

solution, the nonspecifically adsorbed ions are distributed in a three

dimensional region called the diffuse layer, which extends from the

OHP into the bulk of the solution. The structure of the double layer

can affect the rates of electrode processes [53].

A.1.2. Faradaic processes

Electrochemical cells in which faradaic currents are flowing are

classified as either galvanic or electrolytic cells. A galvanic cell is

one in which reactions occur spontaneously at the electrodes when

they are connected externally by a conductor.

Considering a general electrode reaction such as O + ne- ↔ R, it is

possible to consider that the current (or electrode reaction rate) is

governed by the rates of processes such as:

Mass transfer to the electrode surface and vice versa

Electron transfer at the electrode surface

Homogeneous or heterogeneous chemical reactions on the

electrode surface

Other surface reactions, such as adsorption, desorption, or

crystallization (electrodeposition).

The rate constants for some of these processes depend on the

potential. The simplest reaction only involves mass transfer and

electron transfer steps, however, chemical reactions and

adsorption/desorption processes are taking place in some reactions.


Figure 56: Pathway of a general electrode reaction.

The response of the electrode to the current flow corresponds to an

overpotential that is the sum of different terms associated with the

reaction steps occurring with a finite rate in the electrochemical cell:

ηmt (the mass-transfer overpotential), ηct (the charge-transfer

overpotential), ηrxn (the overpotential associated with a preceding

reaction), etc. The electrode reaction can then be represented by a

resistance, R, composed of a series of resistances (or more exactly,

impedances) representing the various steps: Rm, Rct, etc.

Figure 57: Processes in an electrode reaction represented as resistances.

A fast reaction step is characterized by a small resistance (or

impedance), while a slow step is represented by a high resistance

[53].


The kinetic of a system is strongly influenced by the applied

potential, the nature and the structure of the reacting species, the

solvent, the electrode material and adsorbed layers on the electrode.

A.1.3. The Butler-Volmer equation [52]

As reported above the potential of an electrode strongly affects the

kinetics of reactions occurring on its surface. The phenomenological

model of Butler-Volmer is able to relate the reaction rate (current

density) with the applied electrode potential.

The simplest electrode process can be represented as one electron

process:

𝑂 + 𝑒− ↔ 𝑅

It is necessary to consider the thermodynamics and the kinetics of the

electron transfer process. If the kinetics of electron transfer are rapid,

the concentrations of О and R at the electrode surface can be assumed

to be at equilibrium with the electrode potential, as governed by the

Nernst equation:

𝐸𝑒𝑞 = 𝐸0 + 𝑅𝑇

𝑛𝐹ln (

𝑐𝑂

𝑐𝑅)

where the equilibrium potential is related to the standard potential of

the couple O/R, E0, and the surface concentration of O and R. The

standard potential clearly represents the particular equilibrium

potential when the surface concentrations of O and R are equal. If no

current passes through the cell no chemical change can occur; the

surface concentrations must therefore be equal to the bulk

concentrations 𝑐𝑂∞ and 𝑐𝑅

∞. While no net current is flowing and there

is no overall chemical change in the cell, there must be a dynamic

equilibrium at the working electrode surface, for example the

reduction of O and the oxidation of R are both occurring, but the

processes are of equal rate:


𝑖0 = −𝑖 = ��

where I0 is the exchange current density, and 𝐼 and 𝐼 are the partial

current densities for the forward and backward reactions.

The magnitude of the current flowing at any potential depends on the

kinetics of electron transfer. At any potential the measured current

density in given by

𝑖 = 𝑖 + ��

(where 𝑖 is negative); the current density is dependent on a rate

constant and the concentration of the electroactive species at the site

of the electron transfer, the electrode surface, i.e.

𝑖 = −𝑛𝐹��𝑐𝑂 and �� = −𝑛𝐹��𝑐𝑅

The rate constant vary with the applied electron potential according

to the equations of the form

𝑘𝑐 = 𝑘0

exp (−𝑐𝑛 𝐹

𝑅 𝑇𝐸) 𝑎𝑛𝑑 𝑘𝑎

= 𝑘0 exp (

𝑎𝑛 𝐹

𝑅 𝑇𝐸)

where a e c are constants (between 0 e 1 and generally

approximately 0.5) known as the transfer coefficients for the anodic

and cathodic reactions, respectively. For a simple electron transfer

reaction a + c = 1, so that one transfer coefficient may be

eliminated from any equation; 𝑘𝑐 and 𝑘𝑎

represent the heterogeneous

rate constant of the reduction and oxidation reaction, respectively.

At equilibrium (E=Eeq) the net current density is zero and the

potential of the working electrode is given by the Nernst equation.


Taking these information into account it is possible define the

overpotential as the deviation of the potential from the equilibrium

value:

𝜂 = 𝐸 − 𝐸𝑒𝑞

Taking into account the definition of the exchange current density,

𝑖0 = −𝑖 = �� at η=0 it is possible to obtain the Butler-Volmer

equation:

𝑖 = 𝑖0 [exp (𝑎𝑛𝐹

𝑅𝑇𝜂) − 𝑒𝑥𝑝 (

𝑐𝑛𝐹

𝑅𝑇𝜂)]

It is possible to observe how current density varies with exchange

current density, overpotential, and the transfer coefficients.

Depending on the overpotential value, three cases can be

distinguished: the first two applying an high overpotential value,

while the third applying a very low overpotential:

|��| ≫ |𝑖|

𝑙𝑜𝑔𝑖 = 𝑙𝑜𝑔𝑖0 +𝑎𝑛𝐹

2.3 𝑅𝑇𝜂

|𝑖| ≫ |��|

𝑙𝑜𝑔 − 𝑖 = 𝑙𝑜𝑔𝑖0 −𝑐𝑛𝐹

2.3 𝑅𝑇𝜂

𝜂 ≪ 𝑅𝑇 𝑐𝑛𝐹⁄ and 𝜂 ≪ 𝑅𝑇 𝑎𝑛𝐹⁄

𝑖 = 𝑖0

𝑛𝐹

𝑅𝑇𝜂

The first two cases are known as the Tafel equations and are the basis

of a simple method of determining the exchange current density and

a transfer coefficient.

As shown in Figure 58, both linear segments extrapolate to an

intercept of logi0. The plots deviate sharply from linear behaviour as


η approaches zero, because the back reactions can no longer be

regarded as negligible.

Figure 58: Tafel plots for anodic and cathodic branches of the current-

overpotential curve for О + e *= R with a = 0.5, T = 298 K.

Electrochemical techniques [52]

A.2.1. Potential sweep techniques and cyclic voltammetry

The electrochemical behaviour of a system can be obtained through

a series of steps to different potentials with recording of the current-

time curves. Potential sweep techniques, such as cyclic voltammetry,

are widely used to determine the kinetic parameters for a large variety

of mechanisms. This technique permits to obtain the potentials at

which processes occur and to identify the presence of coupled

homogeneous reactions. Although better techniques probably exist

for the determination of precise kinetic data, cyclic voltammetry is

the technique of choice when studying a system for the first time.


Figure 59 shows the potential-time waveforms used for sweep

measurements.

Figure 59: Potential-time profiles for sweep-voltammetry.

Cyclic voltammetry (CV) involves sweeping the electrode potential

between limits E1 and E2 at a known sweep rate, ν; on reaching the

potential E2 the sweep is reversed. On again reaching the initial

potential, E1, there are several possibilities. The potential sweep may

be halted, again reversed, or alternatively continued further to a value

E3. The linear sweep voltammetry (LSV) involves sweeping the

electrode potential between limits E1 and E2 at a known sweep rate,

ν. In both CV and LSV experiments the cell current is recorded as a

function of the applied potential. Cyclic voltammetry techniques are

used to study a system for the first time: it is usual to start by carrying

out qualitative experiments before proceeding to semi-quantitative

and finally quantitative ones from which kinetic parameters may be

calculated.

Voltammograms are recorded over a wide range of sweep rates and

for various values of E1, E2 and E3; observing how peaks appear and

disappear as the potential limits and sweep rate are varied it is

possible to determine how the processes represented by the peaks are

related. At the same time, from the sweep rate dependence of the peak


amplitudes the role of adsorption, diffusion, and coupled

homogeneous chemical reactions may be identified.

A.2.2. Reversible reactions

For a simple reversible reaction where only O is initially present in

solution

𝑂 + 𝑒− ↔ 𝑅

it is possible to record a voltammogram that appear like a steady-state

I vs E curve when a very slow linear potential sweep is applied to

such a system. Figure 60 shows a typical cyclic voltammogram

where it is possible observe that the charge associated with the anodic

process is low compared to the forward reduction process.

Figure 60: Cyclic voltammogram for a reversible process; Initially only O

is present in solution.

This is because throughout most of the experiment, there is a

concentration difference driving R away from the electrode; most of

the product, R, therefore diffuses into the bulk solution and cannot be

reoxidised on the timescale of a cyclic voltammetric experiment.

-i


The peak current density has been found to be proportional to the

concentration of electroactive species and to the square roots of the

sweep rate and diffusion coefficient for a planar diffusion:

𝑖𝑝 = −0.4463 𝑛𝐹 (𝑛𝐹

𝑅𝑇)

1/2

𝑐𝑂∞𝐷1/2𝜈1/2

This is called the Randles-Sevčik equation and at 25°C this reduces

to the form

𝑖𝑝 = −(2.69 × 105)𝑛3/2𝑐𝑂∞𝐷1/2𝜈1/2

where ip, the peak current density is in A cm-2, D is in cm2 s-1, v is in

V s-1, and cO∞ is in mol cm-3.

The reversibility of the system can be tested checking whether a plot

of ip as a function of v1/2 is both linear and passes through the origin

(or alternatively whether ip/v1/2 is a constant). If this is found to be

true further diagnostic tests can be applied, all of which should be

satisfied by a reversible system.

The test of reversibility has to be applied over a wide range of sweep

rates or false conclusions may be reached. A reversible cyclic

voltammogram can only be observed if both O and R are stable and

the kinetics of the electron transfer process are fast so that at all

potentials and potential scan rates the electron transfer process on the

surface is in equilibrium so that surface concentrations follow the

Nernst equation.

The following experimental parameter values can be useful to

characterize the cyclic voltammogram of a reversible process at

25°C:

𝐸𝑝 = 𝐸𝑝𝐴 − 𝐸𝑝

𝐶 = 59𝑛⁄ 𝑚𝑉

|𝐸𝑝 − 𝐸𝑝/2| = 59𝑛⁄ 𝑚𝑉

|𝑖𝑝𝐴 𝑖𝑝

𝐶⁄ | = 1

𝑖𝑝 ∝ 𝜈1/2


𝐸𝑝 is independent of ν

At potentials beyond 𝐸𝑝, 𝑖−2 ∝ 𝑡

A.2.3. Irreversible systems

In the case of the reversible system discussed above, the electron

transfer rates at all potentials are significantly greater than the rate of

mass transport, and therefore Nernstian equilibrium is always

maintained at the electrode surface. When the rate of electron transfer

is insufficient to maintain this surface equilibrium then the shape of

the cyclic voltammogram changes. For such a system at low potential

scan rates the rate of electron transfer is greater than that of mass

transfer, and a reversible cyclic voltammogram can be recorded;

however, the rate of mass transport increases with increasing scan

rates and becomes comparable to the rate of electron transfer leading

to increase the peak separation. Furthermore the peak height is

slightly reduced from that for a reversible system.

For irreversible systems the peak current density can be calculated

by:

𝑖𝑝 = −(2.99 × 105)𝑛(𝐶𝑛)1/2𝑐𝑂∞𝐷𝑂

1/2𝜈1/2

where n is the number of electrons transferred up to, and including,

the rate determining step. The peak current density is proportional

not only to the concentration and to the square root of the sweep rate,

as in the case of reversible system, but also to the square root of the

transfer coefficient. For an irreversible system, Epc is found to vary

with the sweep rate as shown below:

𝐸𝑝𝐶 = 𝐾 −

2.3 𝑅𝑇

2𝐶𝑛𝐹𝑙𝑜𝑔𝜈


where

𝐾 = 𝐸0 −𝑅𝑇

𝐶𝑛𝐹[0.78 −

2.3

2𝑙𝑜𝑔 (

𝐶𝑛𝐹𝐷

𝑘02𝑅𝑇

)]

The shape factor |𝐸𝑝 − 𝐸𝑝/2| is different for the irreversible case and

is given by

|𝐸𝑝 − 𝐸𝑝/2| = 48

𝐶𝑛

𝑚𝑉 𝑎𝑡 25°𝐶

From this equation it is possible to estimate the factor 𝐶𝑛 required

for the estimation of D and k0.


characterize the cyclic voltammogram of an irreversible process at

25°C:

No reverse peak

𝑖𝑝𝐶 ∝ 𝜈1/2

𝐸𝑝𝐶 shifts − 30 𝐶𝑛⁄ 𝑚𝑉 for each decade increase in ν

|𝐸𝑝 − 𝐸𝑝/2| = 48 𝐶𝑛⁄ 𝑚𝑉

A.2.4. Quasi reversible systems

It is quite common for a process that is reversible at low sweep rates

to become irreversible at higher ones after having passed through a

region known as quasi-reversible at intermediate values. This

transition from reversibility occurs when the relative rate of the

electron transfer with respect to that of mass transport is insufficient

to maintain Nernstian equilibrium at the electrode surface. This

change from reversible, to quasi-reversible and finally irreversible

behaviour can readily be seen from a plot of ip as a function of v1/2 as

shown in Figure 61.


Figure 61: Plot of the dependence of the peak current density on the square

root of the potential sweep rate, showing the transition from reversible to

irreversible behaviour with increasing sweep rate c= 0.5.


characterize the cyclic voltammogram of a quasi-reversible process:

|𝑖𝑝| increases with 𝜈1/2 but is not proportional to it

|𝑖𝑝𝐴 𝑖𝑝

𝐶⁄ | = 1 provided 𝐴 = 𝐶 = 0.5

𝐸𝑝 is greater than 59 𝑛⁄ mV and increases with increasing ν

𝐸𝑝𝐶 shifts negatively with increasing ν

Potential step techniques [52]

A complete analysis of any electrochemical process requires the

identification of all the individual steps and, where possible, their

quantification. Such a description requires at least the determination

of the standard rate constant, k0, and the transfer coefficients, A and

C, for the electron transfer step, the determination of the number of

electrons involved and of the diffusion coefficients of the oxidised

ip

v1/2


and reduced species (if they are soluble in either the solution or the

electrode).

For an electrode reaction where only O is initially present in the

solution

𝑂 + 𝑒− ↔ 𝑅

the potential-time profile (Figure 62) can be obtained choosing a

potential E1 such that no reduction of O, or indeed any other electrode

reaction, occurs.

Figure 62: The potential-time profile for a single potential step

chronoamperometric experiment.

Then at time t = 0 the potential is instantaneously changed to a new

value E2, where the reduction of O occurs at a diffusion controlled

rate. Solving the Fick’s 2nd Law with the appropriate boundary

conditions for a planar electrode it is possible to obtain the Cottrell

equation:

|𝑖| =𝑛𝐹𝐷1/2𝑐𝑂

∞

𝜋1/2𝑡1/2


It is possible to observe in Figure 63 (curve a) that the current falls

as t-1/2. A plot of i vs t-1/2 should therefore be linear and should pass

through the origin, and the diffusion coefficient of species O can be

found from the gradient of the straight line.

Figure 63: I-t responses for a potential step experiment. The potential E2

is chosen so that a: the reaction is diffusion controlled, b: the reaction is

kinetically controlled, and c: there is mixed control.

If the standard rate constant for the reaction described above is very

small (or E2 corresponds to a low overpotential for the reaction), a

current time transient of the type shown by curve b of Figure 63 will

be observed. This is because the surface concentration of O does not

change significantly (<1%) due to the imposition of the pulse, and

therefore diffusion does not play a significant role in determining the

rate. The measurement is essentially a steady state one. For the

intermediate situation, where the rates of diffusion and electron

transfer are comparable, the i vs t transient has the form shown by

curve c of Figure 63; the current falls with time but less steeply than

for the diffusion controlled case. Under these conditions the system

is said to exhibit mixed control.

i


Figure 64: A schematic diagram of the time evolution of the concentration

profiles for a species O undergoing reduction under conditions of mixed

control.

The time evolution of the concentration profile for species O under

mixed control is shown in Figure 64. The flux of O at the surface is

simply given by Fick 's 1st Law:

𝐽 = −𝐷(𝜕𝑐𝑂 𝜕𝑥⁄ )𝑥=0

From Figure 64 it can be seen that this flux is greatest at short times.

Indeed at time t=0, i.e. immediately after the potential step is applied,

the flux would be infinite and therefore the current density at time

t=0 (it=0) would be kinetically controlled, and from its value the rate

constant for the electrode reaction could be found. Unfortunately, for

a number of reasons, the value of it=0 cannot be directly measured.

Since it=0 cannot be measured directly it is necessary to resort to an

extrapolation procedure to obtain its value, and whilst direct

extrapolation of an i vs t transient is occasionally possible, a linear

extrapolation is always preferable. In order to see how this should be

done we must first solve Fick's 2nd Law for a potential step


experiment under the conditions of mixed control. The differential

equations to be solved are

𝜕𝑐𝑂

𝜕𝑡= 𝐷𝑂

𝜕2𝑐𝑂

𝜕𝑥2

𝜕𝑐𝑅

𝜕𝑡= 𝐷𝑅

𝜕2𝑐𝑅

𝜕𝑥2

and for the case where both O and R are present in the solution

prepared for the experiment, and the potential is stepped to negative

overpotentials, these must be solved subject to specific initial and

boundary conditions appropriate to mixed control. It has been found

that at short times a plot of i vs t1/2 should be a straight line of

intercept it=0 given by

𝑖𝑡=0 = −𝑛𝐹��𝑐𝑂∞

from which the potential dependent rate constant k can be obtained.

An analysis of experimental transient data can be made in the

following manner: it1/2 is first plotted as a function of t1/2 as shown in

Figure 65.

The horizontal region at large values of t1/2 corresponds to the

diffusion controlled region, whereas the short time data are affected

by kinetics.

With all determinations of reaction kinetics it is usual to perform

experiments over a range of potentials, thus obtaining a range of

potential dependent rate constants.


Figure 65: Analysis of i-t data from a chronoamperometric experiment

corresponding to �� = 10-2 cm s-1, D = 10-5 cm2 s-1, and 𝑐𝑂∞=10-6 molcm-3;

it1/2 plotted as a function of t1/2

�� = 𝑘0𝑒𝑥𝑝 (−𝐶𝑛𝐹(𝐸 − 𝐸𝑒𝑞

0 )

𝑅𝑇)

�� = 𝑘0𝑒𝑥𝑝 (−𝐴𝑛𝐹(𝐸 − 𝐸𝑒𝑞

0 )

𝑅𝑇)

Plots of 𝑙𝑜𝑔�� or 𝑙𝑜𝑔�� as a function of E should therefore be linear,

and from the gradient of these straight lines the values of the transfer

coefficient A and B may be obtained. If 𝑙𝑜𝑔�� (or 𝑙𝑜𝑔��) is not a

linear function of E it is probable that the reaction mechanism is more

complex than had been assumed.

Bibliography

[1] S. Menezes and D. Anderson, «Wavelength‐Property

Correlation in Electrodeposited Ultrastructured Cu‐Ni

Multilayers» Journal of the Electrochemical Society, p. 440–

444, 1990.

[2] B. Huang, K. Ishihara and P. Shingu, «Preparation of high

strength bulk nano-scale Fe/Cu multilayers by repeated

pressing-rolling» Journal of Materials Science Letters, vol. 20,

p. 1669–1670, 2001.

[3] A. Misra, M. Verdier, Y. Lu, H. Kung, T. Mitchell, M. Nastasi

and J. Embury, «Structure and mechanical properties of Cu-X

(X= Nb, Cr, Ni) nanolayered composites» Scripta Materialia,

vol. 39, pp. 555-560, 1998.

[4] M. Demkowicz and R. Hoagland, «Structure of Kurdjumov–

Sachs interfaces in simulations of a copper–niobium bilayer»

Journal of Nuclear Materials, vol. 372, pp. 45-52, 2008.

[5] T. Hochbauer, A. Misra, K. Hattar and R. Hoagland, «Influence

of interfaces on the storage of ion-implanted He in multilayered

metallic Composites» Journal of Applied Physics, vol. 98, pp.

123516-123517, 2005.

[6] H. L. Chum, V. R. Koch, L. L. Miller and R. A. Osteryoung,

«An Electrochemical Scrutiny of Organometallic Iron

Complexes and Hexamethylbenzene in a Room Temperature

Molten Salt» Journal of the American Chemical Society, vol.

97, pp. 3264-3265, 1975.

118 Bibliography

[7] J. Robinson and R. A. Osteryoung, «An electrochemical and

spectroscopic study of some aromatic hydrocarbons in the

room temperature molten salt system aluminum chloride-n-

butylpyridinium chloride» Journal of the American Chemical

Society, vol. 101, pp. 323-327, 1979.

[8] T. B. Scheffler, C. L. Hussey, K. R. Seddon, C. M. Kear and P.

D. Armitage, «Molybdenum chloro complexes in room-

temperature chloroaluminate ionic liquids: stabilization of

hexachloromolybdate(2-) and hexachloromolybdate(3-)»

Inorganic Chemistry, vol. 22, pp. 2099-2100, 1983.

[9] D. Appleby, C. L. Hussey, K. R. Seddon and J. E. Turp,

«Room-temperature ionic liquids as solvents for electronic

absorption spectroscopy of halide complexes» Nature, vol.

323, pp. 614-616, 1986.

[10] C. L. Hussey, «Room-temperature molten salt systems,»

Advances in Molten Salt Chemistry, vol. 5, pp. 185-229, 1983.

[11] J. S. Wilkes and M. J. Zaworotko, «Air and water stable 1-

ethyl-3-methylimidazolium based ionic liquids» Journal of the

Chemical Society, Chemical Communications, vol. 14, pp. 965-

967, 1992.

[12] D. R. MacFarlane, P. Meakin, J. Sun, N. Amini and M. Forsyth,

«Pyrrolidinium Imides: A New Family of Molten Salts and

Conductive Plastic Crystal Phases» The Journal of Physical

Chemistry B, vol. 103, pp. 4164-4170, 1999.

[13] M. Gaune-Escard and K. R. Seddon, Molten Salts and Ionic

Liquids: Never the Twain?, John Wiley & Sons, 2010.

Bibliography 119

[14] F. Endres and S. Zein El Abedin, «Air and water stable ionic

liquids in physical chemistry» Physical Chemistry Chemical

Physics, vol. 8, pp. 2101-2116, 2006.

[15] F. H. Hurley and T. Wler Jr., «The Electrodeposition of

Aluminum from Nonaqueous Solutions at Room Temperature»

Journal of the Electrochemical Society, vol. 98, pp. 207-212,

1951.

[16] O. Hiroyuki, Electrochemical Aspects, Hoboken, New Jersey:

John Wiley & Sons, Inc., 2005.

[17] F. Endres, «Ionic Liquids: Solvents for the Electrodeposition of

Metals and Semiconductors» A European journal of Chemical

Physics and Physical Chemistry, vol. 3, p. 144–154, 2002.

[18] A. Abbott and K.J. McKenzie, «Application of ionic liquids to

the electrodeposition of metals» Physical Chemistry Chemical

Physics, vol. 8, pp. 4265-4279, 2006.

[19] S. Majumdar, P. Sengupta, G. Kale and I. Sharma,

«Development of multilayer oxidation resistant coatings on

niobium and tantalum» Surface & Coatings Technology, vol.

200, pp. 3713-3718, 2006.

[20] M. Zhang, Y. Zhang, P. Rack, M. Miller and T. Nieh,

«Nanocrystalline tetragonal tantalum thin films» Scripta

Materialia, vol. 57, p. 1032–1035, 2007.

[21] N. Borisenko, A. Ispas, E. Zschippang, Q. Liu, S. Zein El

Abedin, A. Bund and F. Endres, «In situ STM and EQCM

studies of tantalum electrodeposition from TaF5 in the air- and

water-stable ionic liquid 1-butyl-1-methylpyrrolidinium

120 Bibliography

bis(trifluoromethylsulfonyl)amide» Electrochimica Acta , vol.

54, pp. 122-128, 2009.

[22] V. Grinevitch, S. Kuznetsov, A. Arakcheeva, T. Olyunina, A.

Schonleber and M. Gaune-Escard, «Electrode and chemical

reactions during electrodeposition of tantalum products in CsCl

melt» Electrochimica Acta, vol. 51, pp. 6563-6571, 2006.

[23] A. Jiang, T. A. Tyson and L. Axe, «The stability of the β phase

of tantalum: a molecular dynamics study» Journal of Physics:

Condensed Matter , vol. 17, pp. 1841-1850. , 2005.

[24] L. Gladczuk, A. Patel, C. Paur and M. Sosnowski, «Tantalum

films for protective coatings of steel,» Thin solid films, vol.

467, pp. 150-157, 2004.

[25] H. Li, H. Nersisyan, K. Park, S. Park, J. Kim, J. Lee and J.H.

Lee, «Nuclear-grade zirconium prepared by combining

combustion synthesis with molten-salt electrorefining

technique» Journal of Nuclear Materials, vol. 413, pp. 107-

113, 2001.

[26] S. Senderoff and G. W. Mellors, «Coherent Coatings of

Refractory Metal» Science, vol. 153, pp. 1475-1481, 1966.

[27] D. Inman and S.H. White, «The production of refractory metals

by the electrolysis of molten salts; design factors and

limitations» Journal of applied electrochemistry, vol. 8, pp.

375-390, 1978.

[28] A. Girginov, T. Tzvetkoff and M. Bojinov, «Electrodeposition

of refractory metals (Ti, Zr, Nb, Ta) from molten salts

electrolytes» Journal of applied electrochemistry, vol. 25, pp.

993-1003, 1995.

Bibliography 121

[29] S. Senderoff, G. Mellors and W. Reinhart, «The

electrodeposition of coherent deposits of refractory metals»

Journal of Electrochemical Society, vol. 112, pp. 840-845,

1965.

[30] V. Van, J. Madejovà, A. Silny and V. Danek, «Niobium

Complexes in Fluoride Melts» Chemical Papers, vol. 54, pp.

137-143, 2000.

[31] B. Gillesberg, J. Von Barner, N. Bjerrum and F. Lantelme,

«Niobium plating processes in alkali chloride melts» Journal

of Applied Electrochemistry, vol. 29, pp. 939-949, 1999.

[32] L. Polyakova, E. Polyakov, F. Matthiesen, E. Christensen and

N. Bjerrum, «Electrochemical study of tantalum in fluoride and

oxofluoride melts» Journal of Electrochemical Society, vol.

141, pp. 2982-2988, 1994.

[33] P. Chamelot, P. Palau, L. Massot, A. Savall and P. Taxil, «

Electrodeposition processes of tantalum(V) species in molten

fluorides containing oxide ions» Electrochimica Acta, vol. 47,

pp. 3423-3429, 2002.

[34] C. Guang-Sen, M. Okido and T. Oki, «Electrochemical studies

of zirconium and hafnium in alkali chloride and alkali fluoride-

chloride molten salts» Journal of applied electrochemistry, vol.

20, pp. 77-84, 1990.

[35] M. Kawase and Y. Ito, «The electroformation of Zr metal, Zr–

Al alloy and carbon films on ceramic» Journal of applied

electrochemistry, vol. 33, pp. 785-793, 2003.

[36] Z. Chen, Y. Li and S. Li, «Electrochemical behavior of

zirconium in the LiCl–KCl molten salt at Mo electrode»

122 Bibliography

Journal of Alloys and Compounds, vol. 509, pp. 5958-5961,

2011.

[37] Y. Wu, Z. Xu, S. Chen, L. Wang and G. Li, «Electrochemical

behavior of zirconium in molten NaCl-KCl-K2ZrF6 system»

Rare Metals, vol. 30, pp. 8-13, 2011.

[38] H. Groult, A. Barhoun, H. El Gallali, S. Borensztjan and F.

Lantelme, «Study of the electrochemical reduction of Zr4+ ions

in molten alkali fluorides» Journal of Electrochemical Society,

vol. 155, pp. 19-25, 2008.

[39] M. Gibilaro, L. Massot, P. Chamelot, L. Cassayre and P. Taxil,

«Investigation of Zr(IV) in LiF–CaF2: Stability with oxide ions

and electroreduction pathway on inert and reactive electrodes»

Electrochimica Acta, vol. 95, pp. 185-191, 2013.

[40] O. Babushkina, E. O. Lomako and W. Freyland,

«Electrochemistry and Raman spectroscopy of niobium

reduction in basic and acidic pyrrolidinium based ionic liquids.

Part I: 1-Butyl-1-methylpyrrolidinium chloride with NbCl5»

Electrochimica Acta, vol. 62, p. 234–241, 2012.

[41] S. Zein El Abedin, H. Farag, E. Moustafa, U. Welz-Biermann

and F. Endres, «Electroreduction of tantalum fluoride in a room

temperature ionic liquid at variable temperatures» Physical

Chemistry Chemical Physics, vol. 7, pp. 2333-2339, 2005.

[42] O. B. Babushkina and S. Ekres, «Spectroscopic study of the

electrochemical behaviour of tantalum(V) chloride and

oxochloride species in 1-butyl-1-methylpyrrolidinium

chloride» Electrochimica Acta, vol. 56, pp. 867-877, 2010.

[43] T. Tsuda, C. Hussey, G. Stafford and O. Kongstein,

«Electrodeposition of Al-Zr alloys from lewis acidic aluminum

Bibliography 123

chloride-1-ethyl-3-methylimidazolium chloride melt» Journal

of the electrochemical society, vol. 151, pp. 447-454, 2004.

[44] C. Fu, L. Aldous, N. Manan and R. Compton,

«Electrochemistry of zirconium tetrachloride in the ionic liquid

N-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)

imide: formation of Zr(III) and exploitation of ZrCl4 as a facile

ionic liquid drying agent» Electroanalysis, vol. 24, pp. 210-

213, 2012.

[45] L. Lutterotti, R. Ceccato, R. Dal Maschio and E. Pagani,

«Quantitative analysis of silicate glass in ceramic materials»

Materials Science Forum, vol. 87, pp. 278-281, 1998.

[46] S. Kuznetsov, «Electrochemistry of refractory metals in molten

salts: Application for the creation of new and functional

materials» Pure and Applied Chemistry, vol. 81, p. 1423–1439,

2009.

[47] F. Lantelme, A. Salmi, B. Coffin, J. Claverie and. Y. Le

Petitcorps, «Electrochemical deposition of niobium on vitreous

carbon and silicon carbide fibres in fused alkali chlorides»

Materials science & engineering. B, Solid-state materials for

advanced technology, vol. 39, pp. 202-207, 1996.

[48] J. Von Barner, E. Christensen, N. Bjerrum and B. Gilbert,

«Vibrational spectra of niobium(V) fluoro and oxo fluoro

complexes formed in alkali-metal fluoride melts» Inorganic

Chemistry, vol. 30, pp. 561-566, 1991.

[49] Z. Alimova, E. Polyakov, L. Polyakova and V. Kremenetskiy,

«The role of fluoride ions in reduction-oxidation equilibria in

124 Bibliography

CsCl-KCl-NaCl-K2NbF7 melts» Journal of Fluorine

Chemistry, vol. 59, p. 203–210, 1992.

[50] S. Zein El Abedin and F. Endres, «Ionic liquids: the link to

high-temperature molten salts?» Accounts of Chemical

Research, vol. 40, pp. 1106-1113, 2007.

[51] A. Ispas, B. Adolphi, A. Bund and F. Endres, «On the

electrodeposition of tantalum from three different Ionic Liquids

with the bis(trifluoromethyl sulfonyl) amide ion» Physical

Chemistry Chemical Physics, vol. 12, pp. 1793 -1803, 2010.

[52] D. Pletcher, R. Greff, R. Peat, L. Peter and J. Robinson,

Instrumental Methods in Electrochemistry, Cambridge:

Woodhead Publishing, 2001.

[53] A. Bard and L. Faulkner, Electrochemical methods, Hoboken:

John Wiley & Sons, 2001.

[54] S. Randstrom, G. Appetecchi, C. Lagergren, A. Moreno and S.

Passerini, «The influence of air and its components on the

cathodic stability of N-butyl-N-methylpyrrolidinium

bis(trifluoromethanesulfonyl)imide» Electrochimica Acta , vol.

53, pp. 1837-1842, 2007.

[55] R. Randstrom, M. Montanino, G. Appetecchi, C. Lagergren, A.

Moreno and S. Passerini, «Effect of water and oxygen traces on

the cathodic stability of N-alkyl-N-methylpyrrolidinium

bis(trifluoromethanesulfonyl)imide» Electrochimica Acta, vol.

53, pp. 6397-64, 2008.

[56] M. Eshed, S. Pol, A. Gedanken and M. Balasubramanian,

«Zirconium nanoparticles prepared by the reduction of

Bibliography 125

zirconium oxide using the RAPET method» Beilstein Journal

of Nanotechnology, vol. 2, p. 198–203, 2011.

[57] Z. Chen, Y. Li and S. Li, «Electrochemical behavior of

zirconium in the LiCl–KCl molten salt at Mo electrode»

Journal of Alloys and Compounds , vol. 509, pp. 5958-5961,

2011.

[58] Y. Wu, Z. Xu, S. Chen, L. Wang and G. Li, «Electrochemical

behavior of zirconium in molten NaCl-KCl-K2ZrF6 system»

Rare Metals , vol. 30, pp. 8-13, 2011.

[59] M. Druschler, N. Borisenko, J. Wallauer, C. Winter, B. Huber,

F. Endres and B. Roling, «New insights into the interface

between a single-crystalline metal electrode and an extremely

pure ionic liquid: slow interfacial processes and the influence

of temperature on interfacial dynamics» Physical Chemistry

Chemical Physics, vol. 14, pp. 5090-5099, 2012.

[60] S. Zein El Abedin, A. Saad, H. Farag, N. Borisenko, Q. Liu and

F. Endres, «Electrodeposition of selenium, indium and copper

in an air- and water-stable ionic liquid at variable

temperatures,» Electrochimica Acta, vol. 52, pp. 2746-2754,

2007.

[61] P. Chen and I. Sun, «Electrochemical study of copper in a basic

1-ethyl-3 methylimidazolium tetrafluoroborate room

temperature molten salt» Electrochimica Acta, vol. 45, pp. 441-

450, 1999.

[62] K. Murase, K. Nitta, T. Hyrato and Y. Awakura,

«Electrochemical behaviour of copper in trimethyl-n-

hexylammonium bis((trifluoromethyl)sulfonyl)amide, an

126 Bibliography

ammonium imide-type room temperature molten salt» Journal

Applied Electrochemistry, vol. 31, pp. 1089-1094, 2001.

[63] A. Ismail, S. Zein El Abedin, O. Höfft and F. Endres,

«Unexpected decomposition of the bis

(trifluoromethylsulfonyl) amide anion during electrochemical

copper oxidation in an ionic liquid» Electrochemistry

Communications , vol. 12, pp. 909-911, 2010.

[64] A. Misra, M. Verdier, Y. Lu, H. Kung, T. Mitchell, M. Nastasi

and J. Embury, «Structure and mechanical properties of Cu-X

(X = Nb,Cr,Ni) nanolayered composites,» Scripta Materialia,

vol. 39, p. 555–560, 1998.

[65] M. Wei, Z. Cao, J. Shi, G. Pan, L. Xu and X. Meng,

«Anomalous plastic deformation in nanoscale Cu/Ta

multilayers» Materials Science & Engineering A, vol. 598, p.

355–359, 2014.

[66] C. Guang-Sen, M. Okido and T. Oki, «Electrochemical studies

of zirconium and hafnium in alkali chloride and alkali fluoride-

chloride molten salts,» Journal of applied electrochemistry,

vol. 20, pp. 77-84, 1990.

[67] P. Chen and I. Sun, «Electrochemical study of copper in a basic

1-ethyl-3 methylimidazolium tetrafluoroborate room

temperature molten salt» Electrochimica Acta, vol. 45, pp. 441-

450, 1999.

Table of symbols

Roman symbols

A Surface area cm2

ci∞ Concentration of the species i in the

bulk solution

mol cm-3

ciσ Concentration of the species i at the

electrode surface

mol cm-3

Cdl Double layer capacitance F cm-2

D Diffusion coefficient cm2 s-1

E Potential of an electrode vs a

reference

V

E0 Standard potential V

Eeq Equilibrium potential V

F Faraday’s constant C mol-1

i Current density A cm-2

i0 Exchange current density A cm-2

ip Peak current density A cm-2

I Current intensity A

I(Q) Scattered intensity -

Ji Flux of species i mol cm-2

s-1

k Rate constant for a chemical

reaction

cm s-1

128 Table of symbols

k0 Standard rate constant cm s-1

K Equilibrium constant

m Moles of electroactive species in

electrolysis cell

mol

M Molar mol/L

n Electron numbers -

Q Charge C

Q Scattered wave vector -

R Gas constant J mol-1 K-

1

R Resistance Ω

t Time s

T Absolute temperature K

x Distance perpendicular to the

electrode surface

cm

Greek symbols

α Transfer coefficient -

η Overpotential V

θ Scattering angle °

λ X-ray wave length -

ν Potential scan rate V s-1

τ Transition time s

Table of symbols 129

Acronims

Ag Silver

AgCl Silver chloride

Al Aluminium

AlCl3: Aluminium chloride

Au(111) Monocrystalline gold

BDD Boron Doped Diamond

[BMP] [TFSA] 1-butyl-1-methylpyrrolidinium

bis(trifluoromethylsulfonyl)imide

C Carbon

CaF2 Calcium fluoride

Cu Copper

CuF2 Copper(II) fluoride

CV Cyclic Voltammetry

EDX Energy Dispersive X-ray

F Fluorine

F- Fluoride ion

FIB Focused ion beam

FLINAK Eutectic mixture of LiF-NaF-KF

IHP Inner Helmholtz Plane

IL Ionic Liquid

IPE Ideal Polarized Electrode

K+ Potassium ion

KCl Potassium chloride

130 Table of symbols

KF Potassium fluoride

K2NbF7 Potassium heptafluoroniobate

K2TaF7 Potassium heptafluorotantalate

K2ZrF6 Potassium hexafluorozirconate

Li+ Lithium ion

LiCl Lithium chloride

LiF Lithium fluoride

LSV Linear Sweep Voltammetry

Na+ Sodium ion

NaCl Sodium chloride

NaF Sodium fluoride

Nb Niobium

NbC Niobium carbide

Nb2C Niobium carbide

Nb6C5 Niobium carbide

NbCl5 Niobium(V) chloride

NbF5 Niobium(V) fluoride

NbF72- Heptafluoroniobate anion

NbF73- Heptafluoroniobate anion

Ni Nickel

O Oxygen

OCP Open circuit potential

OHP Outer Helmholtz Plane

ppm parts per million

http://www.americanelements.com/kshptf.html

http://en.wikipedia.org/wiki/Potassium_heptafluorotantalate

http://www.americanelements.com/kzrf.html

Table of symbols 131

Pt Platinum

[Py1,4] TFSA 1-butyl-1-methylpyrrolidinium

bis(trifluoromethylsulfonyl)imide

Pyr14 Cl 1-butyl-1-methylpyrrolidinium

chloride

R.T. Room temperature

RTIL Room Temperature Ionic Liquid

S Sulphur

Se Selenium

SEM Scanning Electron Microscopy

Si/BDD Silicon coated with boron doped

diamond

Ta Tantalum

TaF3 Tantalum(III) fluoride

TaF5 Tantalum(V) fluoride

TaF72- Heptafluorotantalate anion

Ti Titanium

V Vanadium

XRD X-ray diffraction

Zr Zirconium

ZrCl3 Zirconium(III) chloride

ZrCl4 Zirconium(IV) chloride

ZrCl2O Zirconium chloride oxide

ZrF4 Zirconium(IV) fluoride

http://www.chemindustry.com/chemicals/0521091.html

List of figures1

Figure 1: Nitrogen filled glove box in which all experiments were

performed. ....................................................................................... 29

Figure 2: The three electrode cell used for all electrochemical

experiments. .................................................................................... 29

Figure 3: Metrohm Autolab 302N potentiostat-galvanostat

controlled by NOVA software (Metrohm, Switzerland). ............... 31

Figure 4: Scanning Electron Microscopy (SEM) equipped with

Energy Dispersive X-Ray (EDX) detector (Zeiss, Germany). ....... 32

Figure 5: Cyclic voltammetries of pure ionic liquid recorded on

BDD electrode before and after heat treatment at 125°C (scan rate

100 mV/s). ...................................................................................... 36

Figure 6: Decreasing trend of end current densities during the

electrolysis process indicating that impurities are being extracted

from the liquid. ............................................................................... 37

Figure 7: Cyclic voltammetry before and after pre-electrolysis

process (scan rate 100 mV/s). ......................................................... 38

Figure 8: Cyclic voltammograms of 0.25 M NbF5 and 0.25 LiF in

[BMP][TFSA] ionic liquid at different temperatures (scan rate 100

mV/s). ............................................................................................. 40

1 a Figures from “On the Electrodeposition of Niobium from 1-Butyl-1-

Methylpyrrolidinium Bis(trifluoromethylsulfonyl)imide at Conductive

Diamond Substrates”, Electrocatalysis (2014) 5: 16-22, Permission

requested.

b Figures from Electrochemical deposition of Cu and Nb from

pyrrolidinium based ionic liquid, Thin solid films (2014) 571: 325-331,

Permission requested.

134 List of figures

Figure 9a: Cyclic voltammograms at different scan rates of

[BMP][TFSA] containing 0.25 M NbF5 and 0.25 M NaF at BDD

electrode at 125 °C. .........................................................................41

Figure 10a: Cyclic voltammograms at different scan rates of

[BMP][TFSA] containing 0.25 M NbF5 and 0.25 M LiF at BDD

electrode at 125 °C. .........................................................................41

Figure 11a: Cyclic voltammograms of [BMP][TFSA] containing

0.25 M NbF5 and 0.25 M LiF at BDD electrode at 125 °C obtained

at different stop potentials (scan rate 200 mV/s). ............................42

Figure 12a: SEM micrograph and EDX profile of the electrodeposit

formed potentiostatically on BDD in [BMP][TFSA] containing 0.25

M NbF5 and 0.25 M LiF at potential of -1.5 V for 1 h at 125 °C. ..44

Figure 13a: SEM micrograph of the electrodeposit formed

potentiostatically at BDD in [BMP][TFSA] containing 0.25 M NbF5

and 0.25 M LiF at potential of -1.5 V for 1 h at 125 °C. ................45

Figure 14a: SEM micrograph of the focused ion beam (FIB) cross

section of the electrodeposit formed potentiostatically on BDD in

[BMP][TFSA] containing 0.25 M NbF5 and 0.25 M LiF at potential

of -1.5 V for 10 minutes at 125 °C. .................................................45

Figure 15a: XRD pattern of layer deposited over the BDD substrate

as a function of the scattering angle 2θ. The Rietveld integral profiles

belonging to the different detected phases are shown together with

the indexing information (vertical bars below the XRD pattern). ...46

Figure 16: Cyclic voltammograms of [BMP][TFSA] containing 0.25

M NbF5 and 0.25 M LiF at polycrystalline gold electrode at 125 °C .

.........................................................................................................49


M NbF5 and 0.25 M LiF at polycrystalline gold electrode at 125° C

(scan rate 100 mV/s). ......................................................................49

Figure 18: SEM micrographs and EDX profiles of the electrodeposit

formed potentiostatically on gold electrode in [BMP][TFSA]

containing 0.25 M NbF5 and 0.25 M LiF at potential of -1.5 V for 1

h at 125° C. ......................................................................................50

List of figures 135

Figure 19b: Cyclic voltammograms of [BMP][TFSA] containing

0.25 M NbF5 and 0.25 M LiF at copper electrode at 125° C. ......... 52

Figure 20b: Cyclic voltammogram of [BMP][TFSA] and 0.25 M LiF

at copper electrode at 125° C. ......................................................... 52

Figure 21b: SEM micrographs and EDX profiles of the

electrodeposit formed potentiostatically on copper electrode in

[BMP][TFSA] containing 0.25 M NbF5 and 0.25 M LiF at potential

of -1.5 V for 1 h at 125° C. ............................................................. 53

Figure 22: Cyclic voltammograms of 0.25 M TaF5 and 0.25 LiF in

[BMP][TFSA] ionic liquid at different temperatures (scan rate 100

mV/s). ............................................................................................. 55


M TaF5 and 0.25 M LiF at BDD electrode at 125° C. .................... 56

Figure 24: Values of current density measured at the peak values of

-1.1 V and -1.8 V identified in the voltammograms obtained with

BDD electrode as a function of the square roots of the scan rate. .. 57

Figure 25: SEM micrograph and EDX profile of the electrodeposit

formed potentiostatically on BDD in [BMP][TFSA] containing 0.25

M TaF5 and 0.25 M LiF at potential of -1.8 V for 1 h at 125° C. ... 58


M TaF5 and 0.25 M LiF at polycrystalline gold electrode at 125° C.

........................................................................................................ 59


formed potentiostatically on gold electrode in [BMP][TFSA]

containing 0.25 M TaF5 and 0.25 M LiF at potential of -1.8 V for 1 h

at 125 °C. ........................................................................................ 60


M TaF5 and 0.25 M LiF at copper electrode at 125 °C. ................. 62

Figure 29: Values of current density measured at the peak values of

-0.8 V and -1.7 V identified in the voltammograms obtained with

copper electrode as a function of the square roots of the scan rate. 62


formed potentiostatically on copper electrode in [BMP][TFSA]

136 List of figures

containing 0.25 M TaF5 and 0.25 M LiF at potential of -1.8 V for 1 h

at 125° C. .........................................................................................63

Figure 31: Trend with time of potential measured during

galvanostatic experiments at different values of imposed current

density. Inset shows a magnification of the first 2 seconds of

galvanostatic experiments. ..............................................................64


M ZrF4 and 0.25 M LiF at BDD electrode at 125° C. .....................66


M ZrF4 and 0.25 M LiF at BDD electrode at 125° C (scan rate 100

mV/s). ..............................................................................................66

Figure 34: SEM micrographs and EDX profiles of the deposit

obtained at -1.85 V for 1 h at 125 °C on BDD cathodes in

[BMP][TFSA] containing 0.25 M ZrF4 and 0.25 M LiF. ...............69

Figure 35: XRD pattern of the deposits obtained at constant

potentials for 1 h at 125 °C on BDD as a function of the scattering

angle 2. ..........................................................................................70


M ZrF4 and 0.25 M LiF at polycrystalline gold electrode at 125° C

(Inset shows cyclic voltammogram of [BMP][TFSA] containing 0.25

M LiF at 125 °C, scan rate 100 mV/s ). ..........................................72


M ZrF4 and 0.25 M LiF on polycrystalline gold electrode at 125° C

(scan rate 100 mV/s) .......................................................................72

Figure 38: SEM micrographs and EDX profiles of the deposit

obtained at -1.85 V for 1 h at 125 °C on gold electrode in

[BMP][TFSA] containing 0.25 M ZrF4 and 0.25 M LiF. ...............73

Figure 39: Trend with time of potential measured during

galvanostatic experiments at different values of imposed current

density at gold electrode at 125 °C..................................................74

Figure 40b: Steady-state response recorded in [BMP][TFSA] at

copper electrode at 125° C (scan rate 0.5 mV/s). ............................77

List of figures 137


dissolved Cu at niobium electrode at 125° C obtained at different

scan rates. ........................................................................................ 78

Figure 42b: Cyclic voltammogram of [BMP][TFSA] containing

dissolved Cu at niobium electrode at 125° C (scan rate 100 mV/s).

........................................................................................................ 79


dissolved Cu and 0.25 M LiF at niobium electrode at 125° C........ 80

Figure 44b: SEM micrographs and EDX profiles of the

electrodeposit formed potentiostatically on niobium substrate from

[BMP][TFSA] containing Cu and 0.25 M LiF at potential of -0.8 V

for 1 h at 125° C. ............................................................................. 81

Figure 45: Cyclic voltammograms of [BMP][TFSA] containing

dissolved Cu and 0.25 M LiF at tantalum electrode at 125° C obtained

at different scan rates. ..................................................................... 82


formed potentiostatically on tantalum electrode in [BMP][TFSA]

containing dissolved Cu and 0.25 M LiF at potential of -0.8 V for 1

h at 125° C. ..................................................................................... 83

Figure 47b: SEM micrograph and elemental maps of the Cu/Nb

electrodeposit obtained in dual bath mode on niobium electrode: first

deposit of Cu prepared at -0.8 V for 1800 s, followed by Nb deposit

obtained at -1.5 V for 3600 s. The procedure was repeated twice. . 85

Figure 48b: EDX profile of the Cu/Nb electrodeposit obtained in

dual bath mode on niobium electrode: first deposit of Cu prepared at

-0.8 V for 1800 s, followed by Nb deposit obtained at -1.5 V for 3600

s. The procedure was repeated twice. ............................................. 85

Figure 49b: SEM micrographs of the focused ion beam (FIB) cross

section of the Cu/Nb electrodeposit obtained in dual bath mode on

niobium electrode: first deposit of Cu prepared at -0.8 V for 1800 s,

followed by Nb deposit obtained at -1.5 V for 3600 s. The procedure

was repeated twice. ......................................................................... 86

138 List of figures

Figure 50: SEM micrograph and EDX profile of the Cu/Ta

electrodeposit obtained in dual bath mode on BDD electrode: first

deposit of Cu prepared at -0.8 V for 1800 s, followed by Ta deposit

obtained at -1.8 V for 3600 s. ..........................................................87


electrodeposit obtained in dual bath mode on BDD electrode: first




electrodeposit obtained in dual bath mode on tantalum electrode: first



Figure 53: SEM micrographs of the focused ion beam (FIB) cross

section of the Cu/Ta electrodeposit obtained in dual bath mode on

tantalum electrode: first deposit of copper prepared at -0.8 V for 1800

s, followed by tantalum deposit obtained at -1.8 V for 3600 s. ......89

Figure 54: Schematic diagram of an electrochemical cell. ............96

Figure 55: Proposed model of the double-layer region under

conditions where anions are specifically adsorbed. ........................99

Figure 56: Pathway of a general electrode reaction. ....................101

Figure 57: Processes in an electrode reaction represented as

resistances. .....................................................................................101

Figure 58: Tafel plots for anodic and cathodic branches of the

current-overpotential curve for О + e *= R with = 0.5, T = 298 K.

.......................................................................................................105

Figure 59: Potential-time profiles for sweep-voltammetry. .........106

Figure 60: Cyclic voltammogram for a reversible process; Initially

only O is present in solution. .........................................................107

Figure 61: Plot of the dependence of the peak current density on the

square root of the potential sweep rate, showing the transition from

reversible to irreversible behaviour with increasing sweep rate c=

0.5. .................................................................................................111

List of figures 139

Figure 62: The potential-time profile for a single potential step

chronoamperometric experiment. ................................................. 112

Figure 63: I-t responses for a potential step experiment. The

potential E2 is chosen so that a: the reaction is diffusion controlled,

b: the reaction is kinetically controlled, and c: there is mixed control.

...................................................................................................... 113

Figure 64: A schematic diagram of the time evolution of the

concentration profiles for a species O undergoing reduction under

conditions of mixed control. ......................................................... 114

Figure 65: Analysis of i-t data from a chronoamperometric

experiment corresponding to k = 10-2 cm s-1, D = 10-5 cm2 s-1, and

c𝑂∞=10-6 molcm-3; it1/2 plotted as a function of t1/2 ........................ 116

Electrodeposition of Nb, Ta, Zr and Cu from Ionic Liquid...

Documents

Transcript of Electrodeposition of Nb, Ta, Zr and Cu from Ionic Liquid...